ANNUAL REPRODUCTIVE CYCLE OF THE
MALE POCKET GOPHER (Geomys pinetis)
KATHERINE CARTER EWEL
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
[N PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UN[1'EMI OF FLORIDoA
I am indebted to Drs. David W. Johnston, John H1. Kaufmann, E. G.
Franz Sauer, and Daniel 8. Waard for their encouragement and helpful
criticism during the course of this project. In addition, Dr. John
Thornby offered considerable advice on the statistical analysis of the
data; Thomas Krakauer assisted me several times in trapping; Robert
McFarlane gave assistance in the preparation of photographs; Dr. Gordon
N. Prine provided environmental data from the Agronomy Farm Weather
Station; and Dr. John F. Gerber assisted in the interpretation of soil
Permission to trap on Stengel Field was kindly granted by Mr..
The University of Florida Computing Center provided the facilities
for the statistical analysis.
TABLE OF CONTENTS
ACK~NOWrLEDGIENITS ............................................ ii
LIST OF TABLES ............................................. iv
LTST OF FIGURES ............................................ v
ASSTRACT ................................................ vii
INTRODUCTION~ ................... ................... ............. 1
Pockcet Gopher Environment and Physiological Adiaptations ..... 2
crf;sts of Envirosnment: on Activity Cycle .................... 3
Reproductive Cycles in Pockiet Cophers ....................... 3
The Reproductive Cycle of Geomys pinetis .................... 5
CharacteL; r: i;Ticsj rf feproductive runs.....................
1Instological Lethod ..........ai~a1.............................. 1
Exammattra~:'?;r j ofr. .he Tstis................................... 1
Examianon f theEptilerate............................... 13~
Crniminatinl Jf the Accessor, Glands .......................... 13
~ii- lccn ..*Che C am es in Reproductinle .re.n..................... 12
C clT Ch..............:3,si -pan.te rse................ il
Compncrtson of ilch e iin Reproducci c o.erin ..........ns..
5;lTeIsi:,~ in C:rad '11 .Or~ ures ....... ..................****** 51
Quartcrl:, ,* Sangle ................... ..........................
LI~~cTERAtUE CITED ........................................... 61l
biE;liG.AHIA SKETCH ....................................... 0
LIST OF TABLES
1. Means of mecasuremecnts from male pocket Cophers ........... 18
2. Independent factors in measurements fromn male
pocket goph~ers ................................... 19
3. Numbers of pocket Gophers withl acini at a given stage
of secretary activity in the prostate glands .............. 22
4. Relationships between thle production of sperm
and accessory gland activity .............................. 23
5. Niumlber of males caught in each trapping period ............ 24
6. F ratios from univariate analyses of variance
shlowiny, significance of diferences between means
of reproductive parameters in 15 trapping periods ......... 26
7. Distribution.of significant peaits among the parameters .... 38
8. Results of canonical analysis ............................. 43
9. Mean values of environmental parameters over the
specified periods of time before approximate
conception dates ......................................... 48
10. Sex, age, and reproductive status of animals captured
in each quarterly sample .................................. 50
LIST OF FIGUlRES
1. Criterion for separating mature from im~mature
male pocket gophers ..........,..................,,,........ 9
2. Testicular volume~s throughout the trapping periods ........ 27
3. Diameters of the testis tubules throughout the
trapping periods ..,......,,....................,.......... 28
4. Diam~eters of the Leydig cells throughout the
trapping periods ................,,............,,,........ 29
5. Volumles of Leydig cell tissue throughout the
tr pring periods **....................................... 30
'..';llo-es io the stages of spermatogenesis
rlwl:.u, uc the trapping periods ........................... 32
.~. .11.tir :f the ep~ididymal tubule throughout'
tra~~ r~c ppin,-, period= ...................................... 33
i. I:ll he.ChS~cr in rhle ..ca5 .lar.:i al pc:roci t. (1and
11 r' -.at l_ .tnl t 11- t ;.jrj i rl !i.e r <[,0~ n ..r.. ;,0.1 mi~~~~~air.,u
.'~ tLrl .. .....isur = ...................................... 395
12.~~i~ F".*er.i n.nchl .oil ruel.cure...... ................... .... -
toi their~~ cirl.cur. ................... ................... .... ;
Lr, th,?. c r..~i'.U ............................~.......***** ;1
15. Pcriodls of reproductive activity based on
all projected conception dates ................................ 47
16. PNonthly soil temperatures andl rainfall
during, quarterly samplinge period .............................. 4'9
17. Periods of reproductive activity accordling to
projected conception dates in quarterly samples ............... 52
18. Level 8 testis (100%) ......................................... 65
19. Level 3 testis (430S) ......................................... 65
207. Level 7 testis (430S) ......................................... 65
21. Epididiynal tubules fromn a male
in active reproductive condition (100X) ....................... ;66
22. Epididymal tubules from inactive male (100X) .................. 66
23. Actively secreting acinus from dorsolateral
prostate gland (430X) ......................................... 67
24. Inactive acinus from dorsolateral
prostate gland (430X() ......................................... 67
25. Lumen and part of cell boundary from an active
seminal vesicle (100X) ........................................ 68
26. Inactive seminal vesicle (100X) ............................... 68
Anut; act of I?/.\artAtion~ Pr." e~nt.--i to the iGraduat~e Coiincil
in~ lartial Fulfillmecnt of tri. r.. .ute=,-n.*nt for Eno iDegr.?i of
licCorl of Uilo opphy
iiALE PrjIl.E T ST1COME (0 ra. va',*= g )
ihairmsni Daytd W~. Johnston
;a jor De~partme~nci Zoclogy
Ain jnalyst5 ~.is made~ nf the reprjdructive crcle of th- southr-
easteirn poclrac rgopher, ocoways pIngy~ j i, to detemnine btch the pacttern
o;f reprodu~ction in chts =pec; 8jies an the marlnittude~ of envLO~irorwntal
influ..nces on reproduction in a fossoriail habhitac ncir Gainesville,
Florida. The primary focus of the investieatiion ias the irale repro-
ductive icycl, but Igros; Ineasulrements of the reproductive.- cond~ition
of the. females ucre jlso usedl.
Thr rien c estis, its rpididymis, and the pro*::imal accessory
glans (;seininl vesicles and1 prostate Pland~js) of 76 ale pocket~
gopher s ollected moicnthly Detween Majrch, 1967, and.j July, 196j8, were
sectionecd ond examir~ned histolo;~teally to hsscss the degree of repro-
IJuctiLe acttvity of coch orrgan. A onc-dajy inalysts of varijnce was
per~formdc onh ejch of the Iramasurer~lent made on the various arrans to,
determine if j monthly cycle of activity was present. A principal
components analysic. was also performed to find out what ir~iterelation-
ships might exist btweeun the parameters. Finally, a canonical
analysis related those parametersj whlch showed cyclical difference;
during the trapping period with corresponding soil temperature anid soil
moisture values to compare the Effects of the environmental pa-a-Pters
on the reproductive cycle.
During a subseqluent tw~elve-month trapping, period, qluarterly popula-
tion samples were made in the same study area. Only gross measureme~nts
of the reproductive condition were made on the 60 individuals in this
part of the investigation.
The reproductive cycle in the male pocket pgopher was found to be
closely correlated with soil temperature and to a lesser degree with
soil moisture. Naximum reproductive activity occurs wlhen environmental
conditions create a friable soil, suggesting an increased likelihood
of intraspecific contacts. Very warm external temperatures and lowJ
soil moisture inhibit reproductive activity indirectly, presumably by
restricting activity within the burrow.
Leydig cells show a net decrease in size and number with age,
after an initial increase at puberty. Increased reproductive activity
is correlated with an increase in Leydie cell diameter. Spermato-
genesis was non-cyclic, but did not cease in the entire population
during the study period. The activity of the accessory glands and
the epididymis served as a more reliable index of an individual's
Fossorial rodents are unique among terrestrial mammals in being
isolated from most of the environmental factors that control daily and
annual biological cycles in many other organisms. Typical of these
cycles is reproduction, which may be controlled or influenced in many
mammals by the available food supply, temperature, photoperiod, or
rainfall. In studying reproduction of a fossorial mammal, these ex-
ternal factors must be translated to their effects on the environment
of the burrow. In this investigation, emphasis is given to an analysis
of environmental effects on the reproductive cycle of a fossorial
rodent, the southeastern pocket gopher (Geomys pinetis Rafinesque).
This animal spends almost its entire life underground and is
well adapted to a fossorial existence, as indicated by its relatively
broad, flattened skull, small eyes and pinnae, stout forelimbs, and
elonejted foreclawrs. It: feeds chiefly on roots obtrined from Lts
burrow, js we-ll as on occjsionjl leajvis pulled into the shallower parts
of the burrow. It is adoomi seen above Eround3, and as. exposed to daiy-
Iirght onl, duirinfi short: perious when mjktng fresh moundls. Pr-s~uilably,
pocket gophers are influincedo more by the environmental of the turrow
than by light. Particularly Iimportant influences would include temi-
parature changEs, gas comiposition, and soil characteristics.
Pockect Gop~her Environment and Physiological Adaptations
Kennedly (1964) found that the temperature within the burrow of
Coomvs bursarius in central Texas is more constant than environmental
temperatures, and that the range of burrow temperatures becomes narrower
at increased depths. McNab (1966) noted that temperature changes within
burrows of G, pinetis in Florida are also less extreme than outside
temperatures. Because of the temperature stability in the burrow,
Wilks (1963) reasoned that there might be less selection pressure
towards strict homothermy. Poor thermoregulation has indeed been found
to be characteristic of fossorial rodents (Gunther, 1956; Wilks, 1963;
McNab, 1966). M~cNab (1966) claimed that conductance is facilitated by
a scanty fur coat, an increase in peripheral circulation, and a general
decrease in body size and mean weight in areas of higher soil tempera-
ture. The greater stress of summer temperatures, accentuated by the
burrow's saturated atmosphere (MlcNab, 1966), might still pose a problem;
the pocket gopher could, however, avoid heat loading by retreating into
the deeper, cooler part of its burrow system. From 10 to 20% of the
burrow systems of Thomomys bottae and G. bursarius lie well below the
superficial runways, which are within 12 inches of the surface (Miller,
The oxygen and carbon dioxide concentrations in the burrow system
fluctuate more widely than does temperature. Increased soil moisture
cause; the carbon dioxvide content, which is initially several times
Sreater than the atmospheric concentration,, to rise even more, probably
due to decreased diffusion through the soil pores (h'ennedly, 1904).
As a result, according to m~easurements mlade~ in Florida by Mci:ab (1966),
- 3 -
the carbon dioxidec concentration in the burrow rejches~ as mulcr aj
20 to 35 times the atmospheric concentration. The oxygen concentra-
tion remains within 80 to 95% of the atmospheric concentration.
Effects of Environment on Activity Cycle
Little is known about the influence of soil moisture and soil
temperature on pocket gopher activity. No published data on activity
cycles within the burrow are available, and Vaughan and Hansen (1961)
remarked that the usual indicator of activity -- the appearance of
fresh mounds -- is really only indicative of surface or near-surface
activity. Trapping experience, however, indicated to these authors
and to Wilks (1963) that soil temperature is an important factor con-
trolling the pocket gopher's daily activity cycle.
Miller (1948) found that the amount of burrowing (determined by
the appearance of fresh mounds) is primarily controlled by soil moisture,
the greatest activity occurring when the soil was most friable.
Kennedly (1964) came to the same conclusion, but the correlation seemed
weaker to him.
hcoroducti-e Cycles in Pocket Caphers
Reprodulctive activity In most5 species of poclket Ccphrs is
limited by their usually solitary behavior, although instanes of
multiple captures uithin one burrow jysterm have been reported for nany
speciies of Genomy5 andl Thomowsj (e.e., E~ni-lish, 1932; Vaughan, 19)62;
W~ls, 963 N~ler L96).Both Vau~nnn (1962) jnd Wilks (1963)
observed that these multiple captures werc miost frequent during the
breeding season and usually involved antmals of opposite sexrs.
Uine (1960) listed only one multiple capture in G.. pinltgg, involving
an adult male and a post-partum female. She also cited 14l instances
during one year of collecting in which a pocket gopher was caught in
one trap when the second trap had been sprung. This, however, could
have been accomplished by one animal's blocking one trap before being
caught in the other. In general, it appears that pocket gophers remain
solitary within their burrows most of the year.
Ecological aspects of reproductive cycles have been studied in
several species of pocket gophers. In east Texas, English (1932)
showed that only one litter per year was characteristic of G. breviceps,
whereas Wilks (1963) reported at least two litters per year for G.
bursarius in south Texas. Peak~s of reproductive activity in G.
bursarius were found in December January, and April May, with
some activity in July September. Wilks suggested that temperature,
humidity, and rainfall were interacting to inhibit summer breeding.
C. bursarius in Colorado produced litters only in the spring (Vaughan,
1962). In this case, the breeding season seemed to be timed so that
the young appeared in late M~ay or June at the peak of vegetative
growth and when the soil was most friable. The adults began to come
into reproductive condition in late winter while the ground was still
frozen to a depth of six inches.
According to Miller (1946), T. bottae in California has only one
breeding season in the northern, more mountainous part of its range,
4hercas the breedint, season in the Eouthern part seemrd to extend
throul-h part of~ the uint-r rainy: season to allow two or three larttrs
Per ye'ar. D~ixon (1929) felt that thi; cycle was attuln1 primarily to
the? appearance of the po et gopher's most iLaportant food plants at
the beginning of the season. Hie found that~ in the- central Sar.~ Joaqlinn
Valley, where alfalfa and other green forage crops were available
throughout the summer, the breeding season was longer. Gunther (1956)
also described the breeding season of this species as restricted to the
cool, wet weather of winter and spring in nonirrigated portions of the
range, but being more protracted in irrigated areas. Bond (1946)
suggested that whereas availability of food might affect the sexual
development of a pocket gopher, contact between the sexes at the op-
timum time for reproduction depended more on the friability of the soil.
W~ing (1960) described the breeding cycle of G. pinatis in Florida,
but made no correlations with environmental factors. Pregnant females
were encountered in her collections throughout the year, but they
ippea~red more frequently in early spring and late summer. Probably
some females had produced at least two litters per year, since double
sets of placental scars occurred in seven females collected between
April and July. Her work included a thorough examination of the females
as well as gross measurements of the male reproductive tracts and epidid-
yrjl spe~rm jsmears. Thil males In her samplFes: showed greatest reprodu..-
tive activity' from Jinuar) Auut,:UL ;h?reaS F:mallCS were most arctive'
in bajrch and July Augusat.
Thec Reprod.uctive Cycle o~f CIoomvS CnetLS
Unli;e ither- species of pockce rophers that Inhabit temip:rte~
rergions, G. plngtlj doecs not ;pp'ar to njve a iell-defined br~eeding
cyile thit can be eastly actltruted to enetironmental 'Iff~-ets n'ing
has suse~r: sted that breeding continues~ throughoutr the- yea-r btil not at
the S;Lme intensity. ThrlET are at least four diiffernt ways in which
the male reproductive cycle could be operating to Generate this
patterns 1.) The males may remain in an active reproductive state
all year, but the females come into breeding condition only at
presumably propitious times. 2.) The males do not remain in an
active state of spermatogenesis all year, but instead store sperm
for a certain period of time after spermatogenesis has ceased, mean-
while retaining a propensity toward mating. 3.) The males may show
fluctuations due to environmental conditions, although never completely
losing the ability to reproduce. 4.) Some males may maintain at least
a threshold ability to reproduce whereas others regress significantly;
a well-defined cycle is therefore not maintained in the population.
The purpose of this investigation was to determine which re~pro-
ductive pattern exists for G. pinetis near Gainesville, Florida, and
to examine in greater depth the influence of environmental factors.
MATERIALS AND METHODS
Study Area and Trapping Nethods
At least five male pocket gophers were trapped every month from
Clarch, 1967, through July, 1968, within a five-mile radius of Gaines-
ville, Florida. After July, 1967, all trapping was done at Stengel
Field, three miles WSWJ of Gainesville. This field, containing a
fairly large population of pocket gophers, was characterized by Pensa-
cola bahia grass sod over Arredonda fine sand and was mowed fairly
regularly throughout the trapping period. Part of the field was used
as an unpaved air strip. The animals were captured in Victor gopher
traps which were checked every two to three hours to avoid excessive
decay of the tissues in the dead animals.
In addition, samples of 15 pocket gophers, including both sexes,
were trapped in N~ovember, 1968, and in February, May, and August, 1969,
to obtain quarterly population samples in Stengel Field.
n total of 106 majles and 91 lamTales was obtained for~ analysis in
the preSent trraVCiestigatio
Climlatological data were- obtained from thie Agronomy Farm Ue~athcr
Station, an ar?a one rntle east of' Stevecl Exeldl and quite similarr to
it in soil characceristiCS and rOVOC. The? LO~pcrJaures usedl in the
statistical analyses kere soil temperatures at a depth of four ir~nces
and were averages between the daily maxima and minima. Estimates of
soil moisture were obtained by combining weekly estimates of evapo-
transpiration with total weekly rainfall. Field capacity, defined
as the ability of a soil to hold water against the pull of gravity, is
13.52) by volume for the top two feet of Lakeland fine sand (Hammond
et al., 1967). Harmmond (pers. comm~.) estimated that this value is
probably very similar to the figures for the Arredonda fine sand in
the trapping area and at thle Agron~omy Farm. With rainfall as input
and evapotranspiration as output, soil moisture was calculated for
each week beginning at a time when field capacity had 'Deen obtained.
Evapotranspiration values were obtained frorm measurements that had
been taken daily on two lysimreters at the Agronom~y Farm. Daily out-
put (leachate) was subtracted from input (2000 mm of irrigation
water daily plus any rainfall) to give a rough indication of soil
moisture (Tosi, 1968). This method should be reliable when used on
a weekly basis, and should certainly provide a relative indication of
periodic soil moisture cycles.
Characteristics of Reproductive Organs
Certain standard measurements were taken of all specimens:
L .I l l I I I
51' 70l 100 20030050
Body; Usigiht )~
Fig. 1. Crnteriors for SeparTiCng macure from irrianaturr mole
- 10 -
altered at puberty, and he noted that this wras reflected in a
change in the log log slope of org~an weight vs. body v~eight,
When the testis volume and3 hodyr weight for all -rocket gophers were
comparred on a logarithm~ic scale, two distinct clusters appeared,
separated by a difference of nearly 900 rmm3 in testis volume measure-
ments. Three individuals intermediate between the clusters were
classified as imm.ature males, because measurements of the epididymal
tibules and the accessory glands were infantile.
Reproductive conditions of females were defined by conditions
of the nipples (furred or exposed), the uterus (degree of vascularity
and swelling) and the pubic bone (the symphysis is resorbed at puberty
Sections were made of various reproductive organs. The right
testis, cauda epididymis, and the entire proximal set of accessory
glands (prostates and seminal vesicles) from each male were preserved
in Bouin's fixative. After being embedded in paraffin, these organs
were sectioned at 10 a? and stained with Heidenhain's iron-hematoxylin.
Sections were m~iade from regulria spaced intervals throughout the
length of the testl; and the iepj.d:,idas Srrsal sections of the acces-
sory glands were ipr:;-'ajr: tofi~litat_ trra:irn the tubules. 3:0.e
with enlarged set. o :i,;l nd., i~ ~..Ee sa.ion v::-~ again chosen a: inteir-
vals to .':- art i_, i::a~icuation --.f lirre unbe~.rs of slides.
Examiniti-on of the fe';tic
Several criteria were used to determine the reproductive condition
of the testis. The initial Gross measurements included the lengths of
the two axes of the testis, as though it were a perfect ellipsoid. The
volume was then determined by the equation: V = 4/3 TT ab2~ na"~ being,
the long axis and "b" the short axis.
An average value for testis tubule diameter was calculated for
each individual from ten randomly selected round tubules.
The spermatog~enic values of twenty tubules, randomly selected among
the round tubules in several sections, were also determined according
to the following scale adapted from Roosen-Runge and Ciesel (1950),
with some modifications proposed by Johnson and Buss (1967)1
Level 1. Liccle cell dOfferentxation, no( Ilumen, spar;e germ
'. Immiature ctubule' with fu primary) ;FpermatoC'Cy~es
3. IMo;tl:, primary' pE~rmac~cy:cs, some se~ondary;
;perna~ctecyte fewr spermaicids
4. Ilucrh spermiiatid formiiation
5. Spe~rma~tius unde~rloing clongation buc not~ yet in
6. Sporl~rmaids In b~undlej moringl touardii woule
7. Spcrmal~to.-oca noting from pe'rip~hery to lume~n;
3., SpermatZO'O lininC lunenn
An "icttvc" rst~i; would have i certain proiport-ion of tubule~s In
Levcls 3 8, ;ince the Leve.l 1; tuoulcs revere to Levcl 3 after losing
their spermatozoa. Inactive testes, i.e., those lacking tubules
that are actively producing, sperm, may have Level 1 and Level 2
tubules as well as Level 3 and occasionally Level 4 tubules.
The specific stage of spermatogenesis shown by each animal was
determined accordinG to the following, characteristics
Stage 0. N'o spermatocytes, totally inactive
1. Primary spermatocytes present
2. Secondary spermatocytes present, some spermatids
3. Many spermatids in Level 4
4. Most advanced spermatids in Level 5
5. M~ost advanced spermatids in Level 6
6. Spermatozoa present in Level 7 only
7. 5%L of tubules contain spermatozoa in Level 8
8. 10% of tubules contain spermatozoa in Level 8
9, 15% of tubules at Level 8
10. 20% of tubules at Level 8
Since the Leydir. calls in the intersrtital tissue are believed
to secrete te~stosterone, the njle hormone that controls the activity
of the secondary sex characteristles (Ploom and Faucatt, 1962), those
Ley'die cells with visible? nuclear and cell rre.-branes +ere measured.
both the cell diameter and the nuclear diamete?~r of round cells erer
recorded at 1000X.
- 13 -
The total volulme of Leydip, cells in the testis was obtained by
the following method, after Groome (1940). Using an ocular grid
measuring 1.92 x 10" mm2 at 1000X and focusing through the 10 a
depth of the section, the total number of Leydig cells in several
randomly selected fields was counted. Since the volume of each field
was known (1.92 x 10-6 mm3), the average number of cells per field
was multiplied by the number of field-volumes contained in that
particular testis to get the total number of Leydig cells in the
testis. The average Leydig cell diameter in each animal was used to
calculate the total volume of the Leydig cell tissue in the testis;
this value was divided by the testis volume for a measure of the
percentage of the total volume.
Examination of the Epididymis
The epididymis consists of three regions the caput epididymis,
which contains both the opididymal duct and efferent ducts from the
rete testis; the corpus epididymis; and the cauda epididymis. The
tubule diameter, cell height, and nuclear diameter were each jveraged
over ten tubules in the cauda cpididymiis.
Eiaminatio~n of thr' AcScEsoY TGlandlS
Each ductus deflercns receives secretions from the seminal
vesicles and the prostate flands before joining the ur~ehrj.
Th~e seminal~ vesicle In the younF. animal andl in the inactive
adult is stronely lobulated, but the cavties~ and recesses ejssntiallyg
disappear during active secretion, when lume~Fn e~xpands. The cells in
the recesses appear to be the miost active secretory units (lioare iet al. ,
1930); ten such cells were measured among the sections from each
individual to give an average height, and the relative amount of
secretion present in the lumen of the vesicle (0, 1, or 2) was noted.
If no secretion was found, the rating was 0. A trace amount was
given a rating of 1, and the presence of enough secretion to cause
distension of the vesicle was rated as 2.
The prostate Gland is divided into three sections dorsal and
lateral units, which are considered as one unit here, and the ventral
unit, which is interior and anterior to the others and seems to be
much less extensively developed (Macklin and Macklin, 1963). Ten
cells from acini of each of the two sections were measured and the
relative secretary activity was also recorded as in the seminal
Statistical Analysis of the Data
Correlation coefficients were calculated for all possible pairs
of measurements (including 16 reproductive parameters and four
environmental parameters) by eliminating from the calculation of a
given coefficient those individuals which lacked either or both of
the measurements. A program from the University of Florida Statis-
ticall Progral: Ltocacy, "Interccrre~lation with D:iSsing~ Dt);E"
(UFj5iLO2), computced those coefficients. This correlatton COclit"
cient miacrix served as input for the Bionedical Coilputecr Protram,
"Factor Analysis" (aI0003i). because the correlation coathicetent
were use~d as the input rather than tnc raw data, thts program pro-
ividea only a principal components analysis. Th~e techniques reduce;
the correlation coefficient m~atrix to a smaller num~ber ofT uco~rrelhteu
variables, or factors, which :oupF the mo~lst shv~ilar parlTm-etrs. Eaich
of the original variables is usighted~ accordingg to the stroniteh of
its relationship with all the other variables in each factor. The
most important factor is the first, since it contains the greatest
proportion of the total variance.
Since the trapping was divided into 15 periods, a one-way
analysis of variance was performed on each of the parameters (BMID01V,
"Analysis of Variance for one-Way Design"). The means of those paran-
eters showing significant F ratios at the .05 level of significance
were then tested for significance, again at the .05 level, by Duncan's
N'ew Multiple Range Test adjusted for unequal sample sizes (Steel and
Fariaiit-rs rhowiac jignifica~nt diffirrnc-s (i.., e*,clical chjnCes
ever clve 16-rFonth triapin g perlIod) uiere Frollped togEther ni ccmparod
v-ich a _er t~ of environmental parimeters in i second multivarriate tech-
nique;, canonical anailysi; (EI:)6I, "Canns;ical Anailisir"). dnly the 31
individuslr +ith co.,plete ;ets of lita for the signifcant parameters
were u;sed. Ecr g;ach ind]iv;idul, the alerage soil moi:ture and roil
tenp'?r; tuiC values for thC two-uce;: irnd fourT-i e:-k pr~~iod~ Prei~'ous to
the: datec of captu~r; werr cailcu.lated.j The~~ c-nonical analysisz treated
the- reprodu~cti'.' psrametecrs as one :et of dnca, the encrirorirental
pac~rtmeter; as; another. To linEer functions; of the Cen:ral forms
Y = "1 1 + *** nt nr:l were conlplted for ealch 3et, ILuch thlat 71 and~ Y2
re re ari:.;-jlly con-l a cd. The "a" inlificiints Cive th~e rclativ' e
im~parrance o'f carh payiraer i --.in cal~llaciing cho, co-fficientE Y.
F?ur clinical anialyse3 were rcan, each1 prograrn including the
jsiame reprodu!ICTive :ataII but e~ .it differren cominatiojci n of sort te~n-
perature and soil moisture values. Thus data on soil temperature for
the two wooks prior to capture were paired with the soil moisture for
two weeks prior in one program and four weeks prior in the second
program. The data on soil temperature for four weeks prior were sim-
11arly pairedi with each of the soil moisture sets in the last two pro-
g~rams. Two independent canonical correlations were calculated in each
program, each correlation maximizing a different environmental parameter.
Each successive correlation was less significant than the preceding one.
Glandular activity between the different kinds of glands and in
individuals of varying degrees of reproductive activity were compared
using the chi-square criterion at the .05 level of significance.
Noncyclic Changes in Reproductive Organs
While the intent of this investigation was to study the cyclic
changes in reproductive activity, it was obvious that non-cyclic, or
age-dependent, changes might also be occurring. Table 1 lists the
differences between the means of measurements in mature and immature
animals, but gives no indication of what further variations might occur
after puberty has been reached. Changes in one organ ma3y also be
correlated with changes in another organ or group of organs; this
may be an important factor that is missed by studying the isolated
parameters. The principal component analysis provided a means of
looking at the set of parameters as a whole, rather than individually.
It selected from the entire set of measurements from mature animals
a small number that accounted for most of the intercorrelations of the
Parameter Immature N Mature N
Body Length (mm) 141.0 14 178.9 .61
Body W'eight (g) 106.2 13 214.5 62
Testis Volume (mm3) 219.4 1Li 3282.8 59
Testis Tubule Diameter (u) 66.8 11 160.4 61
Stage of Spermatogenesis 1.1 11 9.3 61
Leydig Cell Diameter (u1) 5.6 11 8.2 59
Nucleus-Cell Ratio of Leydig Cell (%) 84.4 11 67.4 59
Leydig Cell Tissue Volume (mm3) 21.0 11 370.2 55
Leydig Cell Tissue, % Testis Volume 9.9 11 11.3 55
Epididymi~l Tubule Diameter (u) 48.5 9 226.3 52
Epididymoit: Cell Height (u) 11.3 9 22.5 52
Epididymal1 Cell-Tubule Ratio (%) 23.6 9 12.5 52
Seminal Vesicle Diameter (mm) 0.4 7 1.9 52
Seminal Vesicle Cell H-eight (u) 11.5 7 17.5 54
Dorsolateral Prostate Cell Height (u) 9.3 12 17.4 51
Ventral Prostate Cell Height (u) 10.1 7 14.9 36
- 18 -
Table 1. Means of measurements from male pocket gophers
Table 2. Independent factors in measurements from nale pocket gophers
Parameter 1 2 3
Body L~ength .8553" .0717 -.0244
Body lleight .1373 -.0403 .1853
Testis Volume .5449* .1625 .2069
Testis Tubule Diameter .8860 -.0316 .2910
Stage of Spermatogenesis .9322* -.0521 .1934
Ley~dig Cell Diameter -.6497* .1868 .7099*
Nucleus-Coll Ratio of Leydig Cell -.7208* -.4241* -.0321
Leydig Cell Tissue Volume .3342 .8337* .1660
Leydig Cell Tissue, % of Testis Volume -.1238 .9373*' .0600
Ep'ilid)-nal Tubule Diameter .3919 -.0199 .4130*x
Epididymil Cell Height -.1932 -.1202 .1307
Epididymil Cell-Tu'oule Ratio .5357* .1172 .5303*
Seminal Vesicle Diameter .0348 -.2060 .4868t
Seminal Vesicle Cell Height .1755 .0785 .8616s
Dorsolateral F.rastiet Cell Height .1841 .0575 .8621*
'.'.antrol Frrostate Ce;ll H1ight .195 .20 .75,2"
- 19 -
-indiicate ;oot Inrurtant loadtnes
Hiost of the measurements increased in magnitude with age. The
Leydig cell diameter decreased, however, as did the proportion of
the nucleus to the cytoplasm. The loading for Leydig cell tissue
volume may not have been significant, but its magnitude suggested
that this parameter also increased with age. The percentage of volume
taken up by Leydig cells did not increase, however.
Factor 2, which has no relationship to age or to Factor 1,
showed thst Leydig cell tissue volume and the percentage of Leydig
cell tissue in the testis increased with an increase in the proportion
of cytoplasm to nucleus in the Leydig cell. Since Factor 1 did not
associate increase in percentage of Leydig cell tissue in the testis
with age, but did show that the nucleus-cell ratio increased with
age, Factor 2 suggested that there was actually a cyclic increase
and decrease of both tissue and cell volume.
In Factor 3, the Leydig cell diameter increased with glandular
activity in each of the accessory glands that was studied. Leydig
cell diameter was also related by this factor to the epididymal cell
height, which increased disproportionately to the epididymal tubule
diameter with age. This w~as also reflected in the high loading in
Factor 1, indicating that the epididynis cell increased in height in
relation to the tubule size.
The increase in testis volume wras attribulctable primarily to the
increase in :ize oE theI teatis tubules,, is ind~irrtd by th` dacr?je
in pe~rcenltae of Lcydip sell tissue. Th~e high lo!inf on spe-rmico-
5renesir in Fictor 1 Ehowed that active iperm production ion a not
cyclic in the F1_11-,yojur .. aul, but teameis morei or less constant.
Interrelationships among a~ctivity l.vels in the acces.ory r.14ndis
were considered important even though they did not appear In the
factor analysis. The wa~surements of~t activity were all sign~ificarntly
correlated with one another. Ho~uc.? comparison by means of a chi-
square test of the relative amounts of secretion (absent = 0, slight
r 1, substantial P 2) in the acini or vesicles examined showed that
within a single animal there was no significant difference at the .05
level between the proportions of acini (and of secretary units in the
seminal vesicles) at each stage of secretary activity in the various
glands (Table 3). The accessory glands therefore appeared to secrete
as a unit, rather than varying individually.
A comparison of glandular activity was also made between those
adults with active sperm in their testes (Stages 7 15) and those
without active spermatozoa. The proportion of dorsolateral prostate
acini in each of three stages of secretion was not significantly
different between the two groups. Some secretion in the seminal
vesicle also took place in both groups, but there was signillcantly
grreater secretion in those animals with scti~e spermi. These results
are shown In rable .
Cyclic Changes5 in Reproductive organj
Tne 16-mointh total trapping period ujs broken up into 15
separate perious Lasting from one day to as miuch as 31 days, the
average length being 10 days. At least two weeks usually separated
each trapping period from the adjacent ones. The periods are defined
In Table 5.
5 u )
O I ~O
hervI 0 0
ao .au I
.o CIa a 0
UOFI : a) \o
puc I ,Ml I i 10 1
.1. r I I X
ar -r O
ae o N hi
~ r d I ~I II X 1
Table 5. Number of males caught in each trapping. period
Table 6 lists the results of the analysis of variance test run
on each of the reproduction parameters. The F ratios are listed in
descending order of significance, and are starred according to whether
or not differences between the means of each period were significant.
Changes in the Testis
Changes in testis volume throughout the year are shown in Fig. 2.
Multiple range tests showed that the peaks in Periods 3 and 11 were
significant, but the peak in Period 7 vas not significantly different
from the lows in Periods 5, 6, and 9, perhaps because it was a mean
of only two values. It will be considered significantt here, however,
primarily because of the significance of the corresponding peak,
which included a third individual, in the cycle of testis tubule
diameters (Fig. 3). Period 8 also contributed to the first signi-
ficant peak in Fig. 3. The second significant peak, extended from
Period 11 through Period 13.
Changes in Leydig cell diameter are outlined in Fig. 4. There
are significant peaks at Periods 3 and 11. The changes in the ratio
of nuclear diameter to cell diameter in Leydig cells were also tested
and found to be significant. The resulting pattern was essentially
Table 6. F ratios from univariate analyses of va~Riance
showing significance of differences between
mneans of reproductive paramelters in 15
Parame ter F
Epididymadl Tubule Diameter 4.45**r
Testisc Tubule Diameter 4.36"*
Leydig Cell Diameter 4.04**"
Seminal Vesicle Cell Height 3.63**
Epididpzal Tubule-Cell Ratio 3.57**~
Seminal Vesicle Diameter 2.53*
Dorsolateral Prostate Cell Height 2.52*
Leydig Cell Tissue Volume 2.30*
Nucleus-Cell Ratio of Leydig Cell 2.16*
Testis Volume 2.03*
Ventral Prostate Cell Height 2.03
Body iWeighlt 1.63
St2, a :- E :: ;--:-Ii rvatop n3 5i 1 3.
* I 1~
3000 A S N *
1 7 8 9 0 1 *3 4 1
...pin * *
fi 200 ecclr 'c~l5cruiot h rpig pros ~~d in
conne~s ~can vlue*
- 27 -
A M J P P
0 1 2 1 4 1
~~6 1. ~~ueres ofeheccsis ub~ics nro CI-out the rapin GF~rsd*
.14 ~ -in *oncjiea iu
- 28 -
- 29 -
5~~~~~ .* L L ~--I
A I: J J k S Ii ir i F : I J
9- 3 n 1 2 1 ; i
i ~ ,*in Pel
Fi. L. Dil~trs J ~p riJt: cl r~ru* *~~r rapn
period *oi lecncc~m~ jus
_ __I _~
- 30 -
A J J A 3 O N U J E iNi A Mi J
1 2 3 6 5 3 7 9 10 11 12 13 11 15
Fig,. 5. Volumes of Lcydig cell assueL throughout the trapping
per iods. Sol id iinr cornncecs mran valur-:.
~lchiilll~h the F ratio for -r:-ri~l~.~ijierl;sis vj; not iCnllic.Lic, the
cycl: ouclinad bp the nein; in Pir. ~ ;:~-~-I eiJ correspc~nd to thos~ oi
the previous FarJs.l~trrs. The lack of significance here 'r;ls c~~Isec! by
the great amount of variation around each of the 13eans.
Changes in the Epididymis
Changes in th ratio of epididymal cell height to tubule diaraeter
were, as in the Leydig cells, ruirr~r images of the changes in tubule
diameter. Fig. 7 therefore sh~s significant peaks in epididym~l
tubule diameter at Period 3 and Periods 11 and 12. Corresponding
significant losrs occurred in the ratio of cell height to tubule diam-
C~~ins:~ in ~hs ~ccr-_sorv Cl;nds
;he pslk oi the I~C~S~O~y;land n~~u~-;r.eats uire la~; *;11
d~iinA~ rllin cli3sg in the cFiaio~iiis and ~i;cis ~~ra~TI~cer;.
P~rl) i~ai \raj ~enc~allp suiierJ~-J b~ a Ira, relatlv:lg CCntE cuc~'c.
in tile da~Eol~tril prortjce (Fie. 3), tne firr~ p~j:: YaS T~drh?'j at
Pcriol ~, ~hc~~;J t;i~ 9:Lord Fcltij~Cd r~O;r~ pari.~j i to F-riO3 11
i~:- ;~mlnjl :f~i~?~ ccll I-l~llir: (~~c. CI) r~dsh~-l a p~:il! in Fec~od 3
jn2 ;CJiil florrr I'~ci,-l ~ ~h~':L~t'li Ij Th ;Iliinlll ~.csisle Ji~u~letPr
iFiG. T'';, Iio~'~~cr, rlid nG~~ reach ; sirrriiicsne Fca?: un~ll lice in
the ~r~FFir:- pcricdr iih:n it r.~siin~j I~r~e froii Fec~isij 11 th~ou~h
13. Fl;irJ~~s 1; ir~~~ 15 rrl~:_he hi~~ b~~-n irrclu:r~~ ~n this FI-al', tuc r.cre
;iCnillc~nt!y Invr cnan P:riod 13.
7Fig. 6. Values for the staCes of' spermator~npsenehsii~s rughu the
trapping periods. Solid linf connercts mean values.
_ _____ I ___I_~____
I I I L
- 32 -
10 C I
J J An :Lb
3 4 5 67
M A M J
11 12 13 1:. 15
_IX_ __ _I_~~ ~~~_____~__
- 3 -
. 30 C
I 1_.L_. ...' 2I'--L L. .1__ .1 I
.1 1; .1 .r A i O : L J F il
19*?; 10* 8
1 27 3 's : 6 ; r e 103 11
.0 I_ I-
rr lI -
12 13 1.5 15
ite. 7. i.F.i'tacrt- L 5f Cthe el.I;.I;H..yrsa rIL-Ubul. thlrou ~r <.c l
pw ol. . ln enr?: nc \ lu s
- 34 -
14 r; *j
; : ,L
rrjy~in e Prii0
12~. *,C~l lc~e nr o~ltrl pott iad~r u~o~:
thetrppngpeiod. J~d in cnnr~smen ile*
I t i i I I 1 ~ I I I I*
A E, 3 J 5 i :j D F A ~ J*
G I 0 1 2 3 1. 1
12~ -. *p .rlh;n ceimnl~eil tio~hu h
tr~pinG cr ids. oli lin con~ct mcn vauc*
- 36 -
1 2- * 0 1 2 1 4 1
.i *yl: .5ri
D~~-,~rc-; cf ti~e ~i~lll~ '.!iC1~jtJIC~IIOU: ~e
Comparison of Channgs in TF.-lrrbl:tive~ Crgans
with Environmental Featulres
Table 7 demonstrates considerable overlapping of the peaks among
the parameters. The interpretation of this synchrony requires a close
looke at the environmental features.
Fig. 11 is a summary of the monthly averages of soil temperature
max;ima and minima and of rainfall. Fig. 12 shows the average monthly
values of soil moisture, a function of both rainfall and temperature.
Periods of high soil moisture occurred at times of high rainfall and
high temperatures, and at times of moderate rainfall and low tempera-
tures. Figures 13 and 14 were abstracted from the previous two figures
and present the mean of the environmental histories of the pocket
gophers within each period. Only the two-week-prior soil temperature
history and the four-week-prior soil moisture history are shown. The
two main reproductive peaks matched those of soil moisture fairly well,
and corresponded to intermediate periods of increasing soil temperature.
In spite of this visual correspondence, soil moisture showed no
correlation with any one of the reproductive parameters, and soil
temperature was correlated at the .05 level of significance with only
five of the 14 parame~ters. Since the initial analyses tested only re-
lationships bsetween single factors -- a highly artificial sttuntion --
a m~ultivariatc ana~lysi was chosen that would jccounlt for knocractrion
amo'nG the two sets5 of parameIters.
The results of the canonical analysis are shown in Tablz 9. Of
the four proCgress that were run, only che results from the one sho inr:
the highest correlation -- the combination of soil moisture two weeks
prtor to capture and soil temperature four weeks prior to CapcureP --
X X X
X X X XXX
Soil Temper iture (oF)
Z8 R- 8 o.-
. *..**. o
E *M J
(s 4u ) t i
.. Z M
...1 I .._
(*Io;,-O v 2,p;10
- .1 -
1 1 I I 1-1 I 1 I 3-
L .r..??: Ij J k*! i
11 3 .\ : i 7 li. 9 10 11 1? 13 10 ;5
rlig. 13. Averyc~ soit ccannerarture 11Eones~ Of pooril:eC Cop;II.CIrl wantn
ecjo crnpptra, pcrii-! io; tio I:-TE-lo prior to that~ captur e
- l2 -
1-, J J A 5 0 0 i J F : ijJ
1 .' 3 5 6 ; ; 9 13 11 12 13 1; 15
FiC. 11. A-tecr;.:. soil ci;;L tur hlSis ri? J POC poc; It celi.**ra IIChin eCjl h
t~r7ppin; ?erlm] fo~r focur ;ear:~ s prior to the'ir .'Pture.
Table 8. Results of canonical analysis
inae .1M i:,t otatlaIn
- 03 -
Soil Mioisture -- 2 weeks
Soil Temperature -- 4r weeks
Testis Tubule Diameter
Leydig Cell Diameter
Nucleus-Cell Ratio of Leydig Cell
Leydig 7sll Tissue Volume
Ep?did:Tal TubulD Man t*20 :
D~orrseljaterl P'rosta-:- C01 I!11 ht
I:. 2~ ?3.
are listed. The next highest correlation was from the combination
of soil moisture four weeks prior and soil temperature four weeks pri-
or. The canonical correlation coefficients in the second set were
quite close to those in the first, suggesting that there is little
difference in the effects of soil moisture between the trlo- and four-
waeek values. The other twro programs had coefficients significantly
lower than these.
The canonical correlation showed that soil temperature over a
four week period was much more important than soil moisture in
relating the two sets of variables. The second correlation implied
that some of the variables were definitely affected by soil moisture,
but the great difference between the two coefficients indicated that
these second values were not so reliable. No formal significance
test was performed; however, all coefficients greater than 0.3000
were considered highly important.
Not all the reproductive organs seemed to respond with the same
magnitude of change, or even in the samne direction. Three of the
parameters did not show significance in either correlations dorso-
lateral prostate call height, epididym~al tubule diameter, and Leydig
cell tissue volume. For the most part, however, the relationship was
anl inverse? one: the measurements of reproductive organs decreased with
an increase in soil temperature. M~ost of the second set of coeffi-
Immnature Animals and Pregnant Females
Little is known about the gestation and weaning periods of
Gr. pinetis. Howard and Childs (1959) estimated the gestation period
of T. bottae to be 30 days, but Schramm (1961) observed two pocket
gophers (T. bottae) giving birth only 19 days after copulation; their
young weighed 2 3 g. Barrington (1940) found that the weight of
three newborn G. pinetis averaged 5.1 g, and he estimated that the
gestation period in this species is 30 days. Miller (1946) reported
that weaning takes place in T. bottae after 35 to 37 days, and sexual
maturity is reached after three months. Wing (1960), however, indicated
that from the evidence provided by seasonal percentages, G. pinetis
of both sexes take nearly six months to reach sexual maturity.
Wing (1960) described the largest embryo in her collection as mea-
suring 39 mmn from crown to rump. After estimating the gestation
period to be one month and assuming that the maximum crown-rump
length of an embryo is 40 mmr, an approximation of each embryo's age
in days may be calculated. This information can then be used to
estimate the date of conception.
- 46 -
Uei ght Age Weight Age
g months CI months
65 2 t0 2
90 3 80 3
115 4 100 4
140 5 120 5
165 6 140 6
Again estimating the Gestation period to be about one month,
the date at which conception took place can be calculated by adding
one month to the age and subtracting the total from the month of
Fig. 15 shows the distribution of approximate conception dates
for both pregnant females and immature animals over the total trapping
period. Table 9 compares the average soil moisture and soil tempera-
ture histories, averaged over two-week and four-iweek periods, respec-
tively, before each projected conception date in both the immature and
the pregnant female groups.
A; crot.11 rof ;in! oe'reei eocpher 3 ..ms c~irgipe :rin l th. quarc erlyy
i...plFing p-ad;....ij, 15 ircm e4ach penaol. 'ie. 1; -haows th-? [.-.ittern ot"
TDb-l 103 givec :he petrcentage ofi animal3 In eichi sevr grou', age,
plap, n peera rprouctveitse.The llireproductivE rondition ofI
r'.4 fe~n:l 1c Ir bjased on the~ size anJ **.i:-.llarr ity of rth. Ilter s-: mal c
1..:r-c C:r. nderal: to L. inl acci.':' rli!.riJduct:'-: c:.nditic.n If~ th- te,-tes
. I I I (.
snoraowr. 30 .a.D:E*3* M
Table 9. M;ean values of environmental parameters over the
specified periods of time before approximate
conception dates. Overall means are calculated
from capture dates of males throughout the year.
(Itwo-!Neek Average) (Four-!eck Average)
Croup Z-Vol. OF
Immature Animals 9.62 69.78
Pregnant Females 7.24 71.73
Conception Dates 8.67 70.52
Overall M~eans 6.82 + .68 74.24 + .14
- 8 -
Soil T-unneccturer ('':
o~ ~ a
o ~ ~ ;c O\
::::: :::: :-
:Cn ut v u E
Table 10. Sex, age, and reproductive status of animals captured in
each quarterly sample
Grouo 1 2 3 4
Percent Immature A~nimals 7 13 20 60
Percent of Adult Famales
which are Pregnant 0 28 50 0
Percent of Adult M~ales and Females
in Active Reproductive Condition 36 85 73 33
11: 11/9/68 11/12/68
21 2/20/69 2/23/69
31 5/26/69 6/5/69
4r 8/11/69 8/13/69
were enlarged andi dark~, and if the epididymal tubules were easily
visible to the naked oye.
The projected conception dates for both immature animals and
pregnant females were combined and are shown in Figl. 17.
1. j _5 1 ...L .-
suspeu.:D .0le~yu.; pemps
- 52 -
DISCUSSION AND CONCLUSIONS
The reproductive organs of the male pocket gopher show seasonal
changes that are related to changes in soil temperature and soil mois-
ture. The possibilities that males maintain active spermatogenesis
all year or that cessation or fluctuations in spermatogenesis occur
synchronously throughout the population were, therefore, eliminated.
Instead, it appears most likely that while some periods seem optimnum
for all male pocket gophers, unfavorable periods do not cause abridge-
ment of reproductive activity to the same degree in all the males in
the population. In reaching this conclusion, much insight was gained
into age~-dependentllc changes. Particularly valuable was information
provided on changes within the testis.
Even after sexual maturity is reached, the testis volume, for
example, was found to increase wlith age. This would be due to two
factors increases in testis tubule size and in interstitial tissue.
Testis tubule diameter also increased with age, but the increase in
voiljume of the cocal Lejidi :011 c;useu provedFI ntI to bC ;1Loniituant.
TheC Per~:CJnCar of Leydxglt cCll Cissue voaI~TCe toi coalj ':ILB volum i sred o
agec. Tih.: diineter ofth clLeyic;~'i c 11 nucleus also de~~crat ed sitatft-
iantly with~ ar;e r:lanee to the~ Jia~iter of thei entir c;ell.
Thil; situation sugg2ert that eh.. raEE ofE Leyail,: cell d~ivltionI
ia;cel-re~rat at pulb-rty, accom~pant~sa bi a disprspjrtio atce incriase in
the Crouth rite of ths c;tjplam relstive to chi nucl~:us. Atot
- 54 -
the post-pubertal Leydie cell is larger than the pre-pubertal cell,
its diameter decreases with age. Conaway (1959) recorded a similar
phenomenon in the eastern mole (Scalopus aquatius) The infantile
testis was characterized by abundant but small Leydig, cells with very
little cytoplasm. Early increases in testis volume were due primarily
to increases in Leydig cell volume until testis tubule enlargement
and the concurrent beginning of spermatogenesis became significant
at puberty. After this stage, the diameter of the Leydig cell decreased
by one-third, increasing again slightly at the end of the active
breeding period. The cell diameter was found to be largest during the
period of sexual inactivity.
In studies on the rat, Clegg (1966) found that the number of
Leydig cells actually declined following puberty. He suggested that
the accessory reproductive organs, which showed their maximum growth
acceleration at that stages were more sensitive to the androgens being
produced by the Leydig cells.
Albert (1961) indicated that Leydig cells in bulls also decreased
in size from 5 to 15 years, after an initial increase in both number
and size fromt two years to five years. According to the same source,
the number of Leydig cells in man declines wiith age.
The percentage of Leydig cell tissue volume in the hippopotamus
testi; decreased fromii 641 in rthe ver ;ou~n5 jnimit to 32' at putercy,
Th~e djJcrdi se in size ofl Lcydir cell: in paclkac oph'ri 1E r:ll
n;t an unumuail situationn. Facter r ?rel.tedl th:~ increase in L;:dry
cell diame~ter wiith7 th: ri3 in acti.;ety Of th4 accessor gla'5ndl, a .
well as ~ichi the increase in the dia.-eterr of' the epidid:.-ilal tubules;
relative to the epidid),nal :?ll hei::hts. UnlikeG th2 siri~tuaon des--
cribed earlier in molrr (Counavay, 1959), bol~rilonel SecCrtll on 15 PremnT-
ably associated with both a tempo.r3ry increase in diameter of the
Leydig cells as well as an increase in the ratio of cytoplasm to
nucleus. Enis evidence, tog~eher vith the associations in Factor 2
which herer nentioned earlier, leaves little doubt that a cyclic
increase aind delCeasle of both Leydig cell diameter and of the tissue
volume is imposed on the long-term pattern of diminution.
The height of the epididymal cell was found to increase in
relation to the tubule diarieter with age. The cell height was also
a~ssociiccd un~h Lcydi2 ~cll diise-ltar in factor 2. Ine utterr rasponre
CO;.~rteslewi Co th2 ojbse~Vations sIrwar~izeaC by :13nn (1965), which
showed that~ Jcr~ital of rperm tr. an cpididysiiras sever: friomi che testi;
was J.2pe'r.:?nt on th? continued pr:senze of che testis in thie tody. He
:onclu:ed that~ the condition of the~ epithe~lium in the opididynall tubule
4.15 itrcn,_ly influenc~d by the male Eev bcrcoone. Th: cfunction oC thle
Iplid-id..sh I'snn ;urnised, ii* th?r:for not ojnly to Nerve as j
rap,strry for. speCrm but also: toj produc? a sccretrion san:h-:iru cffc-
the~ in yreservinee or parhalps naturintg thr rperm.
Well-defIneaL1 cy.cles durinL thi saud~ periost were found onily to
the tesicis :O~lume, L-.yd-iS :?1 c ll it6anret andl cissu: volumeo, and
opl~igl.-altutal-1di:ietr in cel heght 15 stage of sp:r..atogen-
.sir **ai t.18:-.1y depe:ndent~ on ie, th2 .arijticen in stages~ :u'Ong the
iiket Coiler: 1.1 aCllh Ltr'Pping pe~iod b-:-in: suffiici:nt to reject
rto possibility of there taken a ;ynchronized populationn cycle in the
prods:C,-.io n of :90tn;-~tlro Jea 41terllehl dcfin tE~ cycles v~rre dcctected in
- 56 -
the activities of the seminal vesicles and the dorsolateral prostate,
they were less distinct than the cycles mentioned earlier. Moreover,
activity in the prostate glands did not necessarily cease when sperm
were not being produced. The seminal vesicles, however, showed a pro-
nounced increase in activity in animals with active sperm production.
Price and iiilliams-Ashman (1961) found that testicular hormones pro-
vided most of the control over activity of the accessory glands. They
also concluded that the chief function of these glands is to secrete
th~e seminal plasma.
Internal control of the reproductive system in the male pocket
gopher seems to rest at the hormonal level. Maximum size of the
Leydig cells reached two significant peaks which corresponded to
similar peaks in size of the epididymis and accessory glands. The
lack of complete correspondence is likely attributable to a low
threshold in glandular response to testosterone. Van Tienhoven (1968)
has also suggested that maintenance of secondary sex characteristics
outside thle breeding season may be the result of androgen secretion
by extra-testicular sources such as the adrenal cortex.
Strongly influencing hormonal control, however, is environmental
control. In the canonical analysis, most of the measurements taken
of thre reproductive organs decreased in rmaginitudel or otherwise
denoted less activity, under conditions of high soil temperacture and
- 57 -
Soil moisture is latecally t~h? lesr iimprrcanc: jf the Cjto calli
it seldom dips below 507. of the field capacity, and therefore shows
comparatively little variation. The distribution of G. pinetis may
well be limited by soil moisture, since this species is found only
in the well-drained sandy soils in the Southeast (McNab, 1966). The
possibility mentioned earlier that excessive Soil moisture might
prompt a pocket gopher to dig to the surface, and perhaps more ex-
tensively underground as well (fMiller, 1948), may be valid, but
apparently this response does not serve as the sole cue for timing
the reproductive cycle.
Temperature does affect the cycle strongly, most likely in an
indirect fashion. Because the largest p~roportion of the burrow lies
near the surface of the ground, it may be inferred that this is where
the greatest amount of digging takes place. The pocket gopher,
however, is a poor thermoregulator and therefore cannot long with-
stand the heat stress of diggeing in a shallow, warm burrow. Gunther
(1956) proposed a tehivi.-ral mechanism in which the gopher makes use
of deeper, cooler Ilace of its burrow system in the summer, plugging
some of the shallow feeding tunnels. The diminished amount of digging
r~ltini froml this ciJsatio'n of acci.:ity r..uld PreS1Lalyjl le-dj to
d~innIshedI Intraspelciifc contacc. Thiis rculatlon of acci\-c would
Effecove.ly restrict mosti rcFpretterie ictivity to cerratn times~ of
Dil r .r_luictiv c:le Cive:~ nor avide~nce of tsing .triiCtly Cjimed.
Is:rnac~Ojeneslt, for inrIcance, iras four.1 to be n)n-cy'clir ac least
dur~ing th-is rrat) dy priod. .4 nacre carefu'll Ltudy of thc repFroducltiv
tri.e itself mighe rrecal :.0000i C; ChF spenO a: some pliCe alongi
- 58 -
the ductus deferens. The great variation in sperm production at any
one time convincingly suggests that the reproductive cycle is not a
strict endogenous rhythm, but is influenced by environmental fluc-
tuations and could be triggered by the right circumstances -- i.e.,
favorable environmental conditions and the presence of a receptive
female -- at any time of the year. Since there was a low level of
reproductive activity that continued throughout much of the study
period, at least some random encounters in the "off season" might
Probably little selection pressure operates to channel repro-
duction into a strictly timed cycle. Food is certainly most abun-
dant in the spring months but is still available, particularly in a
grassy area such as Stengel Field, in sufficient quantity throughout
the year so that a newborn gopher would not starve. Contact between
pocket gophers of the same sex frequently results in harmful injuries
to one or both animals (Mliller, 1964; Vaughan, 1962). Such contact,
if it should occur between a male and female,raight also be harmful
if neither were in active reproductive condition. It would obviously
be advantageous to a pocket gopher to be capable of mating should an
indjtivdual. OE~ RO o;pposite 3p:; be enconjrllt:tered. PIO1;eveTr, Fnding:
cvery; ct-r ed maintaining~ the ~crcpruc-tiver r: t~ In an -acti'a state~
dulcti\ .:Criv'ity..ilnn me:imuni cotacat is I p:i cte:d, bu~t rellhionoes
ch cr.tir: populjtioon cOcrcc- to 2n in:ctile j~3re jt jch,;-r tim,?r.
The reproductive cycle in the male pocket gopher (Geomys pinetis)
is closely correlated writh soil temperature and to a lesser decree
with soil moisture, but does not seem to be strictly controlled by
either of those parameters. Very rarm external temperatures and lowl
soil moisture appear to restrict activity within the burrow at certain
times of the year, and according;ly reduce intraspecific contacts.
Maximum reproductive activity corresponds to those environmental con-
ditions that create a friable soil.
Internal seasonal changes in the reproductive organs and glands
are presumably in turn effected by hormones, particularly testoster-
one, which is secreted by the Leydig cells. Leydig cells in adult
pocket gophers 'are larger in diameter and more numerous than those in
immature animals. A not decrease in both size and number of Leydig
cells occurs after puberty, but the cells still increase in diameter
at times of increased reproductive activity, suggesting increased
normaon11 production. This cyclic change is correlated with increases
In use1 hcllp~h of thei ;rrEarcto., epithd~l~uns in thec jcces~sory Il.nd:,. and
Spe;;;c ~rmatognest has foun-1 to rbe norn-c,.clic" during; the ;tudyl: riod.
Although ch? ncan :p~-~~ernatop e iclu?; for ijch pcrto.1 foillowed
c:.cili pjttern~ simlsr t.:- the on.: cullined by the L~yplc cell axanat-:C,
th1 varixtenn jbout the mean: uali tao rgreat to illou~ SjtlatiCtIc al 1-
ntfican~ce jprerm Fgraduction Ehrougho~~u t the3 ropullation thor'fre con-
tinuedl at least at a threshold level throughout: the study period, and
reproductive activity at the organ level was restricted more by the
condition of the accessory glands and the epididymis than by the
availability of sperm.
Albert, AZ. 1961. The mammalian testis, p. 305-365. In HJ. C. YounC
(od.), Sex and internal secretions. Vol. 1, 3rd ed. W~illiams &
Hilktins Co., cat~iLnaiCi
BarrinCgton, B. A., 1940. The natural history of pocket rgophers. Ml.S.
Thesis, Univ. of Fla. 194 p.
Bloom, ii. and D. WJ. Fawcett. 1962. A textbook of histology. Wi. D.
Saunders Co., Philadelphia. 720 p.
Bond, R. M!. 1946. The breeding habits of Thomonmys bottae in Orange
County, California. J. hNammL~al. 27: 172-174.
Clegg, E. J. 1966. Puoortal growth in the Leydig cells and accessory
rieprod!uit ti.. G;r;;ins of the~ rat. .T. Anat. Tr0: 367-370.
5Anatoy, 1 1'.'. 193"*. I--5.~. l ...uJactr i n .jE l of th~iE IEaster,\l mll .; lrh3
Engllirh, P. F. 192;-.2. Some habitsic c the~ pocket ,7.oph-ir, IEco-c.3 L~cc:-teep
Groom1, J. F-.. 19-au. Th n;usanahl me-olificatter, oi the interjtitril
tlizzc of th-: cath in the fraJc tat (rcropos).j) Proc Co. 3~.;:..
London 1101 j;-;L.
Gunt~h-:F ,1. li. 1G56. 5tudlie on th~e nile rep-rodu.:clt'. s..ac.1m of the~
nirdergrrou~ 3nil ahlt brriersj onl wats. Fctantion andl me...r~ne nto
Lakla~ind fine sind. 5011 and Crop Lo.:. of Fla., Proc~. T's 11-1C'.
Hua r, F. L. 1923j. The~ absor~pt;io of th-; p:-to ......phys;id of thle pC3. etr
3oph*:r I6000-*3 burySciusl i;hau). nlmer. :'jt. 55: 93-G6.
Howard, W. L. jJnd fl. E. rhild~s, Jr. 195-). EcloC:.' of pa:!:et poph~rj
writh emphjSIj on ThirmemEL I-trrj icuj. Hilgardia 29: 2??-jMC.
- 62 -
Johnson, 0. 18. andi I. 0. Duss. 1967. The testis of the African ele-
phacnt (Loxodocnta africana). I. Hlistoloygical featu~res. J. Reprod.
Fort. 13: 11-21.
Ken~nerly, T. E3. Jr. 1964. Microenvironmental conditions of the pocket
eopher burrow. Tex. J. Sci. 26, 395-441.
Laws, R. NI. and G. Cloug-h, 1966G. Observations on reproduction in the
hippopotamus H~iopopotamus amphib~ius Linn., p. 117-140. In I. W.
Rowlandls (ed.), Comnparative biology of reproduction in mammals.
Zool. Soc. of London, Academic Press.
Macklin, C. C. and M. T. Ma!cktlin. 1963. The seminal vesicles, prostate
and bulbourethral Glands, p. 1773-1822. In E. V. Cowdry (ed.),
Special cytology. Vol. III, 2nd ed. Hafner Publ. Co., N'ew York.
M~ann, T. 1964. The biochemistry of somen and of the male reproductive
tract. 2nd ed. John Hdiley & Sons, Inc., New York. 493 p.
Mic~ab, 8. K. 1966. Thie metabolism of fossorial rodents a study of
convergence. Ecology 47: 712-733.
Miller, Mi. A. 1946. Reproductive rates and cycles in the pocket gopher.
J. EMamm~al. 27: 335-358.
.1948. Seasonal trends in burrowing of pocket gophers (Thomomys).
J. M~ammal. 29: 38-44.
Miller, R. S. 1964. Ecolog~y and distribution of pocket gophers
(Geomyidae) in Colorado. Ecology 45s 256-272.
M~oore, C. R., 5. Price, and T. F. Callagher. 1930. Rat-prostate
cytology as a testis hormone indicator and the prevention of
castration changes by testis-extract injections. Amner. J. Anat.
Price, D. and Hi. G. W~illiams-Ashman. 1961. The accessory reproductive
glandls of mammals, p. 36G-448. In i. C. Young (ed.), Sex and inter-
nal secretions. Vol. I, 3rd ed. Williams & W~ilkins Co., Baltimoore.
Roosen-Runge, E. C. and L. 0. Giesel, Jr. 1950. Quantitative studies
on spermatogenesis in the albino rat. Amler. J. A~nat. 871 1-30.
Tosi, J. A. 1968. C1culo del balance hidrica, p. 42-48. In 2-01, J. J.T
and A. Madriz, Zonas de vida de Venezuela. Republican !de '":r-~~:nel ,
M~inisterio de A~ricultura y Crfa, Direcei6n de Investigaci6n.
Van Tienhoven, A. 1968. Reproductive phy~siology of vertebrates. I. B.
Saunders Co., Philadelphia. 498 p.
Vaughan, T. A. 1962. Reproduction in the plains pocket gopher in
Colorado. J. M~ammal. 43: 1-13.
,and R,. MI. Hansen.. 1961. Activity rhythm of the plains pocket
gopher. J. Mlammal. 42: 561-543.
!lilks, B. J. 1963. Some aspects of the ecology and population dynamics
of the pocket gopher (Geoms burarius in south Texas. Tex. J.
Sci. 15: 241-283.
Nting, E. S. 1960. Reproduction in the pocket gopher in north-central
Florida. J. Mammlal. 41: 35-43.
Fie. 18. Level 8 tests (100X)
Fig. 19. Level 3 testis (430X)
Fig. 20. Level 7 tesjtis (430%~)
- 66 -
Fig. 21. Epididymal tubulcs from a male in active
rcproductive condition. The lumiina are
filled with sperm. (100X)
* Fig. 22. Epididymal tubules from inactive male (100X)
- 67 -
Fig. 23. Actively secretinp: acinus from dorsolateral
prostate Gland (430X)
Fi~g. 26. Inactive acinus from dorsolateral prostate
Fig. 25. Lumen and part of cell boundary from an active
seminal vesicle (100X)
Fig. 26. Inactive srminal vesicle (100X)
Kathorine Carter Ewel was born September 30, 1944, at Clens
Falls, New York. During the summer of 1961, she attended Cornell
University on a National Science Foundation Summer Fellowship, and
Graduated from Clens Falls H~igh School in June, 1962. In June, 1966,
she received the degree of Bachelor of Arts with a major in zoology
from Cornell University. During the summer of 1966, she participated
in a field biology course at Tulane University on a USDHENI Environ-
mental Training Crant and in the fall of 1966 enrolled in the Craduate
School of the University of Florida. She worked as a teaching assis-
tant in the Department of Zoology until Arugust, 1967, and was awarded
an NDEA Title IV Fellow~ship from September, 1967, through August,
1969. During the summer of 1968, she was enrolled in the tropical
biology program at the Universidad de Costa Rica under an N'SF Graduate
Research Fellowrship awarded by the Organization for Tropical Studies.
She is currently employed: as a Temporary Instructor in Zoology at Duke
University while completing her work toward the degree of Doctor of
Philosophy at the University of Florida.
This dissertation was preparedl under ther direction of the
chairman of the candtidate's supervisory commoittoo andi has been
approved by~ all memlbers of that commnittee. It wras submitted to
the Dean of the College of Arts and Sciences and to the Craduate
Council, and was approved as partial fulfillment of the reqluire-
ments for the degree of Doctor of Philosophy.
Dean, Co~llee of Arts S~l sciences
D an, Graduate Schoo