QUARTERLY JOURNAL OF THE FLORIDA ACADEMY OF SCIENCES
DEAN F. MARTIN, Editor BARBARA B. MARTIN, Co-Editor
Volume 51 Spring, 1988 Number 2
DESICCATION AND CRYPTIC NEST FLOODING AS
PROBABLE CAUSES OF EGG MORTALITY IN THE
AMERICAN CROCODILE, CROCODYLUS ACUTUS,
IN EVERGLADES NATI(-O': \ PARK, FLORIDA'
FRANK J. MAZZOTT1(', JAMES A. KUSHLAN(2 AND ANN DUNBAR-COOPERI3'
i'Graduate Program in : : The Pennsylvania State 7-. University Park,
Pennsylvania 16802; (2Department of'. i... 1 Sciences, East Texas State University,
Commerce, Texas 75428; .*: .. ...: of .. and : .... :, RSMAS,
The University of Miami, Virginia Key, Florida 33149
ABs rFACr; .r and desiccation probably caused .. :: of eggs of the American Croc-
odile (Crocodylus acutus) in Everglades National Park, Flooding was subterranean with no sign
evident above ground. Apparent desiccation occurred in a year (1981) of" .-- ... low rain-
fall. The timing of nesting seems to be ::., scheduled, with the developmental period brack-
eted by . and .. conditions. The success of this strategy is shown by the
:, ..' low rate of .. ..... most years.
THE status and population dynamics of the American crocodile (Crocody-
lus acutus) have long been a matter of concern and controversy. Ogden (i .
concluded that the number of American crocodiles in the United States was
not increasing, and that an important factor regulating their numbers was
mediocre nesting success, caused primarily by embryonic mortality.
Most crocodile nesting activity in Everglades National Park occurs near
Florida Bay in an area circumscribed around Little Madeira and Joe Bays
(Moore, 1953; Ogden, 1978; Mazzotti, 1983; Kushlan and Mazzotti, i'
The nesting season occurs in March through ..:::: i and is divided into nest
preparation (beginning of March to mid-May), egg-laying (mid-April to mid-
May), development (mid-April to mid-August), hatching (mid-July to mid-
August), and post-hatching (mid-July to end-of-August) periods.
SMazzotti: Department of a ' Range Sciences, University of Florida, 3245 College
66 FLORIDA SCIENTIST [Vol. 51
Crocodiles deposit eggs in cavities dug in soil along creek banks or exposed
shorelines. The clutch of . is subsequently covered for the duration of
development. Conditions within the nest cavity may adversely affect the
growth and development of crocodile embryos. Ogden (1978) suggested that
embryonic mortality was probably caused by low nest temperatures in marl
substrate. The environment of nest cavities of crocodiles in Florida Bay has
been documented by Lutz and Dunbar-Cooper (1984). The nest temperature
regime they found 1::::: their study did not i :i developing embryos. How-
ever, they hypothesized that embryonic death could result from desiccation.
As part of their study, gas and temperature probes were placed in three croco-
dile nests during 1980. While taking a sample of gas, water was withdrawn
from the middle of the clutch, .-... -- the first evidence that subterranean
flooding was occurring in crocodile nests. The purpose of the present paper is
to describe flooding and desiccating events as probable causes of hatching
failure in nests of the American crocodile in Florida.
METHODs-Crocodile nests were located by searching shorelines and creek banks carefully by
boat, helicopter, and on foot. Nest sites were characterized by morphometry (mound or hole) and
by substrate (marl or sand). To provide an ..1 criterion for morphometry, the elevation of
the top of the substrate above the egg cavity was compared to that of the .. area. If
there was no difference in height, the nest was considered to be a hole nest. The elevation of the
bottom of the clutch was measured relative to Florida Bay or creek water levels during 1980.
We defined hatching failure to be the lack of successful emergence from the egg. Hatching
failure could be the result of : : or embryonic i Embryonic . !., may be
Table 1. Nesting of American crocodiles in Florida during 1980 and 1981. Mean clutch size
was 39.3+9.1 (SD), n=8 nests.
Elevation of Number of Number of
bottom of eggs lost eggs presumed
Clutch Nest Nest Nest clutch above to lost to
number morphometry location substrate sea level (cm) : desiccation
1 Mound Shore Sand 72 0 0
2 Mound Shore Sand 64 0 0
3 Mound Shore Sand 45 0 0
4 Hole Creek Marl -22 40 0
5 Hole Shore Sand 6 14 0
6 Mound Shore Sand 25 0 0
7 Hole Creek Marl 24.5 0 0
8 Hole Shore Sand 14 6 0
9 Hole Shore Sand 0 22
10 Mound Shore Sand 0 0
11 Mound Shore Sand 0 10
12 Mound Creek Marl 0 0
13 Hole Shore Sand 35 0
14 Hole Shore Sand 0 9
15 Hole Creek Marl 0 0
16 Mound Creek Marl 0 0
17 Hole Shore Sand 0 13
MAZZOTTI, ET AL.-CROCODILE EGG MORTALITY
-" 5 -25W
0II i 0 F
MAR APR MAY JUNE JULY AUG SEPT
Fic. 1. Rainfall and changes in water levels measured at the Little Madeira Bay environmen-
tal station during the 1980 ....... ... Similar trends were found at the other environmental
caused by predation or by environmental factors. Egg fertility was determined by the presence of
external -,,. on the egg shell (Ferguson, 1982). The stage of development of dead embryos
was determined by opening and examining eggs that failed to hatch.
After the first clutch was discovered to be flooded in 1980, the water level within the :
was monitored every two to four days for the duration of flooding, by measuring the elevation of
ground water in a small pit dug next to the nest. Water salinity was measured with a tempera-
ture-compensated refractometer (American ., :i Co.) Water levels and salinity in the creek
adjacent to the clutch were also monitored .....': ........ Flooding of three other nests was
discovered during 1980 and 1981.
( : .. in surface water level were measured bi-weekly at five locations in the nesting area.
Rainfall was measured at the same locations by a direct reading rain gauge. Water level fluctua-
tions were measured on a meter stick fixed to the bedrock. The change in water level was plotted
relative to the lowest water level recorded during 1979-1980.
Rainfall records (U.S. Department of Commerce) were obtained for two ', ,i .. sta-
S.... ...; ", area. A 25-year period of record, 1957-81, was used to determine the rainfall
for the incubation period.
i*-*. . AND DiscussION-Twelve clutches were found in i : .: and nine
in 1981. Eight of the 1980 clutches were located prior to hatching and are
discussed here. Mean clutch size was 39.3 eggs (SD 9.11, n= 8). Eighty-nine
percent of eggs laid in these eight clutches were fertile. In no clutch were all
eggs fertile. Location, morphometry, and substrate of each clutch are de-
I in Table 1.
Rainfall in southern Florida is seasonal, with 60- of the annual rain-
fall occurring from May through October. During the wet season the rainfall
is bimodal, with peaks occurring early and late in the wet season (Fig. 1).
The onset of the rainy season coincided with egg 1 : i:;,,
Water levels in northeastern Florida Bay fluctuate both seasonally and
over shorter time periods. These fluctuations are caused by rainfall, evapora-
tion, and wind. Lunar tides have little effect in this area (Tabb et al., i: ,
No. 2, 1988]
68 FLORIDA SCIENTIST [Vol. 51
Water level changes in the study area, as represented by the Little Madeira
Bay station, during the 1980 nesting season are shown in Figure 1. Although
short-term fluctuations occur, water levels generally increase during the incu-
bation period ". T..iy) and peak after hatching (Fig. 1).
F1 ..' ... ,, '. :.l. destroyed one clutch in 1980 (4), one in 1981 (:
and ,.. ;. .i destroyed two clutches in 1980 (5 and 8) 1i .i.i 1). Flooding
events were closely monitored for clutch 4 during 1980 (Fig. 2). The eggs
were flooded twice, yet no above-ground water was observed 1'-- :: either
episode. During the first flooding, at least 60% of the egg cavity was flooding
with 36 ppt sea water, and ,.. !. water levels may have occurred before moni-
o .--- Creek
S20- \ ,
> 2- Top of
LU \ // Clutch
lo- \ //
10 12 14 16 18 20 16 0 1 18 2022 24 26 28
FIG. 2. Water level changes in a crocodile clutch (4, Table 1) during 1980, showing flooding
MAZZOTTI, ET AL.-CROCODILE EGG MORTALITY
touring began. Ground water level in clutch 4 during the second flooding
episode was 15 cm higher than during the first flooding event. The salinity
was lower and the entire clutch was submerged during the second flooding
episode ._ 2). The rainfall peaks in June and July, 1980 1 1) occurred
close to the flooding episodes.
During the first flooding episode, two other clutches situated at low eleva-
tions (5 and 8) were examined. At this time the bottom of clutch 5 was 7 cm
above ground water while clutch 8 was 12 cm above ground water. Clutches
5 and 8 were not checked during the second flooding episode. Later examina-
tion of eggs that failed to hatch in clutches 5 and 8 indicated that egg mortal-
ity probably occurred during the second flooding event.
......; of flooding and embryonic death could be determined by an ex-
amination of the eggs. Little development was observed in clutch 4. i:
higher in the egg :v were slightly more developed (larger embryos) than
lower in the cavity. This observation .. . that mortality occurred at
different times with the lowest eggs dying first. All of the eggs examined from
clutches 5 and 8 had embryos in a more advanced developmental stage than
eggs from clutch 4. Clutches 5 and 8 also contained at various develop-
mental stages, with the most developed embryos occupying the top of the
clutches. This pattern of differential loss of eggs from flooding was also ob-
served by Webb et al. (1977) in Australian Crocodylus porosus,
i .... as a cause of embryonic mortality has been documented for
:/ .: mississippiensis (Joanen, 1969; Metzen, 1977; Goodwin and
Marion, 1978; Dietz and Hines, 1980; Kushlan and Kushlan, i::'--.,, for
.. . niloticus (Pooley, : ',, for : ........... ... (W ebb et al., .
Magnusson, 1982; Webb et al., 1 :.: :. for Crocodylus johnstoni (Webb et al.,
1983); for Caiman - .....- (Staton and Dixon, 1977); and for Caiman
(Crawshaw and Schaller, 1979). This is the first time nest flooding has
been documented for the Crocodylus acutus, and the first time that flooding
was.'. .. .. .' to be subterranean or cryptic, with no surface water present.
The duration of submergence was not determined .: -. 1 in this study,
but twelve hours of submergence in fresh water is the critical tolerance limit
for the eggs ofA '' : :* .. ..."' .. : (Joanen et al., 1977) and Crocodylus
.:. f .:, (Magnusson, 1982). Given such a short critical submergence time,
and, because of the absence of standing water, it is obvious that flood-in-
duced embryonic .......! .:. can occur in American .. .....: 1 nests yet not be
documented by observers monitoring the fate of the eggs.
Lutz and Dunbar-Cooper (i :-. concluded that with the exception of
flooding events, nest moisture depended upon rainfall. Rainfall during the
1981 incubation period was the lowest amount of rainfall for the 25-year
period of record at both the Royal Palm (E-.: _1 National Park) and
Tavernier, Florida stations.
Packard and co-workers (1982) stated that although rigid-shelled eggs of
reptiles are usually insensitive to the hydric environment, there may be some
S::: :: if the nest is very dry. ...... (::' has found that the presence
No. 2, 1988]
70 FLORIDA SCIENTIST [Vol. 51
of air chambers in alligator eggs is caused by excessive water loss. Tracy
(1981, pers. comm.) has found, also in alligators, that incomplete yolk me-
tabolism is caused by excessive water loss from .-:: due to dry nest substrate.
If we assume that these relationships are also true for the American crocodile,
54 eggs from four nests failed to hatch in 1981 apparently due to desiccation
(Table 1). After a nest was abandoned by the adult crocodile, the unhatched
eggs were opened and examined .. :.- .-, ... of these eggs had air sacs and
lacked developed embryos. The other 27 had full-term embryos with incom-
pletely absorbed yolk sacs. There was no evidence of embryos in eggs with air
sacs. Thus it appears that low rainfall during 1981 apparently caused desic-
cating nest conditions, resulting in egg mortality.
Modha (1967, for r.......-........ niloticus) and Staton and Dixon (1977, for
Caiman .* ... .7.. found that eggs in nests in desiccating substrates had low
hatching success. In addition to loss of eggs, newly hatched crocodiles in 1981
looked different than in previous years. They had swollen abdomens and
umbilical scars that had barely closed enough to encompass the remaining
yolk. Also, as predicted by Packard et al. (l^-,. the apparently desiccation-
stressed hatchlings measured in 1981 were smaller in mass and snout-vent
length (Kruskal-'-.' .:i test, p<0.001) than similarly measured hatchlings in
1978-1980 (Mazzotti and Kushlan, unpub. obs.). An alternate explanation is
that egg mortality was not caused by desiccation per se but rather by elevated
temperatures in dry nests. However, abnormalities of embryos and !. ..!:...
caused by high temperatures, such as kinked tails (Bustard, 1971; Webb et al,
1983) were not observed, further supporting the interpretation that desicca-
tion, and not high temperatures, killed the developing embryos.
In Florida, Crocodylus acutus nest in the spring, avoiding the low tem-
peratures of winter and high i : i. : :'-::. of late summer. This timing also
avoids the dry season, when desiccation is a threat, and peak water levels,
when flooding is likely. On this schedule only extremely wet or dry years
cause the level of embryonic mortality described here. In a more unpredicta-
ble environment in northern Australia, Magnusson (: *. suggested that the
prolonged breeding season of Crocodylus porosus was the result of the unpre-
dictable timing of floods. In Florida a more predictable seasonal cycle of
rainfall at the northern limits of the range of the American crocodile ..
selection pressure for marked seasonality in nesting.
Distribution of. ,:, --. -.- mortality (Table 2) was non-random between
sand and marl (X2 test, p<0.01, n=17) and mound and hole nests :.7 test,
p<0.05, n= 17). Flooding of hole nests was the most frequent cause of egg
mortality. The discovery of cryptic nest flooding was the most significant
finding of this study. Currently this level of embryonic mortality is low in
most years (Mazzotti, unpub. obs.), however, an important implication of
cryptic nest flooding is that low elevation hole nests, like those along creek
banks, may be subject to increasing failure due to flooding, caused by high
fresh water discharges into the northeastern Florida Bay mangrove zone.
Fresh water discharge can be caused by local rainfall or water releases result-
MAZZOTT1, ET AL. -CROCODILE EGG MORTALITY
ing from upstream water management practices. Increased loss of creek nests
would be detrimental to the population of American crocodiles in Everglades
National Park, since most of the surviving hatchling crocodiles come from
creek nests (Ogden, 1978; Kushlan and Mazzotti, 1988b).
ACKNOWLEDGMENTs-This study was supported by the U.S. National Park Service. We thank
W. B. Robertson Jr., P. Patty, P Lutz and T. Jacobsen who worked with us on various aspects of
this study. C. Tracy and M. Ferguson generously provided us with their ..-- 1'.1i.h-ed observa-
tions. This paper is part of a dissertation submitted in partial requirement for the degree of doctor
of "** .... i.n .. : Pennsylvania State University i i':. W. Dunson, G. Kelly, D. Pearson and
T. .:ii ar ,Lr.:. : i reviewed drafts of this paper.
BUSTARD, H. R. 1971. Temperature and water tolerances of -.uihali'. crocodile eggs. Brit. J.
CRAwsHAw, P. C., AND C. B. SCHALLER. 1979. Nelrii, 'Paraguayan caime .. ."an ui..', in
Brazil. Report No. 11 to IBDF. 19 pp.
DIETZ, D. C., AND T. C. HINES. 1980. Alligator nesting in northcentral Florida. Copeia.
FERcJUSON, M. W. J. 1982, The structure and composition of the .:. i" and membranes of
Alligator mississippiensis. Trans. Zool. Soc. Lond. 36:99-152.
GOODWIN, T. M., AND W. R. MARION. 1978. Aspects of the nesting ecology of American alligators
(Alligator mississippiensis) in northern Florida. Herpetologica. 34:43-47.
JOANEN, T. 1969. Nesting ecology of alligators in Louisiana. Proc. Ann. Conf. Southeast Assoc.
Game Fish Comm. 23:141-151.
__ L. McNEASE, AND G. PERRY. 1977. Effects of simulated flooding on alligator eggs.
Proc. Ann. Conf. Southeast Assoc. Fish Wild. Agencies. 31:33-35.
KUSHLAN, J. A., AND M. S. KUSHLAN. 1980. Water levels and .::,.M,,i nesting in the Everglades.
Proc. 2nd Conf. on Sci. Res. in Nat. Parks. pp. 8-16.
KUSHLAN, J. A., AND F. J. MAZZOTTI, 1988a. Historic and present distribution of the American
crocodile in Florida. J. Herp. in press.
1988b. Population biology of the American crocodile. J. Herp. in press.
LUTZ, P., AND A. C. DUNBAR-COOPER. 1984. Nest environment of the American crocodile. Co-
MAGNUSSON, W.E. 1978. Nesting ecology of Crocodylus porosus, Schneider, in Arnhem Land,
Australia. Ph.D. Dissert., University of Sydney, Sydney, Australia.
1982. Mortality of eggs of the Crocodile Crocodylus porosus in northern Australia. J.
MAZZOTTI, F. J. 1983. The ecology of Crocodylus acutus in Florida. Ph.D. dissert. The Pennsyl-
vania State University, University Park, PA.
METZEN, W. P. 197 ::. i mri et,:.l.-' f alligators on the Okeefenokcc National Wildlife Refuge.
Proc. Ann. ( .) ... ..,th.. c. Fish Wild. Agencies. 31:29-32.
MODHA, M. L. 1967. The ecology of the Nile crocodile (( .. ':.. ,l*,.. *. Laurenti) on Central
Island, Lake Rudolph. Afr. Widl. J. 5:74-95.
MOORE, J. C. 1953. The crocodile in 7 '..J .e National Park. Copeia. 1953:54-59.
OGDEN, J. 1978. Status and nesting biology of the American crocodile, Crocodylus acutus (Rep-
tilia: Crocodilidae) in Florida. J. Herp. 12(2):183-196.
PACKARD, M. J., G. C. PACKARD, AND T. J. BOARDMAN. 1982. Structure of egg shells and water
relations of reptilian eggs. Herpetologica. 38:136-155.
POOLEY, A. C. 1969. Preliminary studies '. ie hr.-: .. of the Nile crocodile dyl niloti-
cus in Zululand. Lammergeyer. 3.:', '22-h3
STATON, M. A., AND J. R. DIXON. 1977. Breeding biology of the spectacled Caiman crocodiles, in
Venezuelan Ilanos. U.S. Dept. Interior. Fish Wildl. Serv. Wildlife Research Report 5. 21
TABB, D. C., D. L. DuBRow, AND R. B. MANNING. 1962. The ecology of northern Florida Bay
and adjacent estuaries. Florida State Board Consn. Techn. Ser. No. 39. 81 pp.
No. 2, 1988]