The Moving Median Method to Estimate Normal Egg Parameters for Alligator Populations in the Southern Everglades, Southwest Louisiana, and Lake Griffin
IAAAM 1999
P.T. Cardeilhac1; D.L. Winternitz1; J.D. Barnett1; O.L. Bass2; W.F. Wolff3
1University of Florida College of Veterinary Medicine, Gainesville, FL, USA; 2National Park Service, Everglades National Park, Homestead, FL, USA; 3Department of Biology, University of Miami, Miami, FL, USA


The environmental status of a lake or area of habitat is known to affect reproductive parameters of its alligator population.1,2,3,4,5 Therefore, an annual evaluation of the reproductive status of the alligator population in an area like the Southern Everglades (Everglades National Park) is also a continuing indication of environmental status and quality of this habitat. The determination of reproductive potential and egg parameter values for this population can be done with a minimum of disturbance; while the capture and evaluation of a large sample of adult alligators from the area may not be practical or desirable. In our study, reproductive potential and normal egg parameters were determined and control charts made from data obtained during a 4-year study of the Everglades National Park, an 11-year study of Southwest Louisiana (Rockefeller Refuge) and an 11-year study of Lake Griffin in Central Florida. Our primary objective was to determine reproductive potential and normal egg parameter values in order to establish control charts for the study areas. The Rockefeller Refuge was considered to be a protected, stable, highly productive control area. Lake Griffin was included in the study because it underwent severe environmental changes between 1994 and 19981 and an understanding of the effect of environmental changes on reproductive potential and egg parameter values is needed. Reproductive potential and egg parameters were characteristic and relatively constant for the Southern Everglades and Southwest Louisiana over the study period; however, changes that occurred in the environmental status of Lake Griffin in 1997 and 1998 caused significant deviations from the normal values established between 1988 and 1994. The data suggested that changes in values might have been caused by the exposure of the alligator population to a hepatic toxin.2

Materials and Methods

Random samples of 90 to120 eggs were collected annually for 4 years (1995 through 1998) from 30 to 35 different nests at the Everglades National Park in the Southern Everglades. The area studied is approximately 500,000 acres. Sample size was estimated to be approximately 12.5% of the eggs in each nest based on a determined average annual clutch size in the area of 24 eggs. The number of nests sampled represents 75% of the approximately 40 nests projected for the study area for a normal year. All eggs were incubated for 14 days and those that failed to produce a detectable band (attached chorioallantois) were designated infertile (or had suffered very early embryonic death). Infertile eggs were used to determine egg parameters. Samples of infertile eggs from each of a minimum of 20 clutches were collected each year from the Rockefeller Refuge in Southwest Louisiana over an 11-year study period from 1988 to 1998. Eggs were collected from 41 to 52 nests on Lake Griffin in 1988, 1993, 1994, 1996, 1997 and 1998.2 Eggs collected in 1996, 1997 and 1998 were considered out of statistical control and were not used in the determination of normal values. The following measurements and calculations were made on infertile eggs: egg length; egg width; egg weight; egg volume; calculated egg volume; volume error (% deviation of measured volume from the calculated volume); length/width ratio; shell weight; % of the egg represented by shell; shell thickness; shell density; yolk weight; % of the egg represented by yolk; membrane weight; % of the egg represented by membrane.1 Descriptive statistics including an examination of the distribution of values for each egg parameter for each year, stem-and-leaf plots (or a horizontal bar chart), box plots and normal quantile plots were made using the SAS statistical package. Median values for egg parameters were determined for each year of the study period and the data examined to confirm that the samples were described by the same distribution. Median plots and moving median plots of 3 were made to estimate the value for each parameter and the plots examined for trends. (Figs 1,2,3) A mean and standard deviation was determined from the moving median values and control limits were defined as 2 standard deviations from the mean value. The mean value was considered to be the "normal" value for the study area.

Results (Table 1 & Figs 1, 2, 3)

Mean values and their standard deviations are given in Table 1. Data for the Southern Everglades were collected for 4 years, Southwest Louisiana for 11 years and 3 usable samples (1988, 1993 and 1994) were collected from Lake Griffin over an 11-year period.

Table 1. Mean values and their standard deviations of egg parameters for Southern Everglades, Southwest Louisiana and Lake Griffin.





Egg length (cm)

7.25 (0.02)

7.26 (0.04)

7.57 (0.01)

Egg width (cm)

4.079 (0.04)

4.20 (0.03)

4.47 (0.04)

Egg weight (grams)

68.85 (0.46)

72.4 (1.47)

86.0 (0.5)

Egg volume error




Length/width ratio

1.766 (0.008)

1.721 (0.009)


Shell weight (grams)

7.39 (0.16)

7.59 (0.14)

8.89 (0.16)

% of the egg represented by shell

11.14 (0.09)

10.57 (0.30)

11.0 (0.05)

Shell thickness (mm)

0.430 (0.017)

0.433 (0.010)


Shell density

2.14 (0.04)

2.12 (0.02)


Yolk weight (grams)




% of the egg represented by yolk

48.0 (0.29)

44.0 (1.88)

48.0 (0.29)

% of the egg represented by membrane

1.06 (0.06)

1.08 (0.04)

1.06 (0.06)

No significant trends were found over the study periods for Southwest Louisiana or the Southern Everglades but trends were observed in Lake Griffin for the years 1996, 1997 and 1998. These years for Lake Griffin were not used in determining normal egg parameter values for egg length, egg weight, length/width ratio, eccentricity, yolk weight, % yolk, shell thickness and shell density. Most of the parameters for Lake Griffin were considered to be out of statistical control for the years 1996, 1997 and 1998 as is shown for yolk weight (Fig 3).

Click on an image to see a larger view.

Figure 1.

Figure 1. Control chart for egg size in Southwest Louisiana over 10 years.

Figure 2.

Figure 2. Determined mean values of shell weight for 6 populations.

Figure 3.

Figure 3. Control chart of yolk weight for Lake Griffin 1988 to 1997.


In the present study, egg parameters remained relatively constant or in statistical control over a 4-year period for the Southern Everglades, over an 11-year period for Southwest Louisiana and over a 6-year period (1988 to 1994) for Lake Griffin. However, for Lake Griffin, egg length, egg width, egg weight, egg volume error, shell weight, % of the egg represented by shell, shell density, yolk weight, and % of the egg represented by yolk moved distinctly out of either the upper or lower limits of the control chart during the years 1996, 1997 and 1998. An example of a control chart for Lake Griffin (yolk weight) is shown in Fig 3. Reproductive potential for the lake in 1997 had decreased to 2.7% of its 12-year average prior to 1995 and there was a sharp increase in dead adult alligators found on the lake in 1997 and 1998 by the Florida Game and Freshwater Fish Commission. It seems reasonable to assume that severe environmental changes occurred in the lake between 1994 and 1996. The first quantitative data indicating possible environmental changes were the reports on declining reproductive potential and changes in alligator egg parameters for the lake.1 Early embryonic deaths, increased yolk size, decreased shell %, increased shell density and decreased hatchling vigor2 were believed to be consistent with the presence of a hepatic toxin in the adult female alligator population. The presence of the toxin was first detected in 1996 but probably entered the system between 1994 and 1996 (Fig 3).


This research was supported as a part of the ATLSS project of the Dept. of the Interior (USGS) and the National Park Service. The authors wish to thank Dr. Ruth Elsey of the Louisiana Department of Wildlife and Fisheries, the Florida alligator farmers, the Florida Game and Freshwater Fish Commission and especially Mr. Arnold M. Brunell, Mr. Lindsey Hoard and Mr. Allen R. Woodward for providing their information and assistance. Mr. Wayne T. McClellan and Mr. Joseph P. Cardeilhac provided technical assistance.


1.  Cardeilhac PT, D Winternitz, JD Barnett, OL Bass, WR Wolff, 1999. The American alligator as a sentinel for the environmental status of southern wetlands. Proceedings of the International Association for Aquatic Animal Medicine 30 (this issue).

2.  Cardeilhac PT, D Winternitz, JD Barnett, E Froelich, KO Foster, JD Ashley. 1998. Declining reproductive potential of the alligator population on Lake Griffin in central Florida. Proceedings of the International Association for Aquatic Animal Medicine 29:30-36.

3.  Woodward AR, HF Percival, ML Jennings, CL Moore. 1993. Low clutch viability of American alligators on Lake Apopka. Florida Scientist 56:52 63.

4.  Mason GR. 1995. Environmental influences on reproductive potential, clutch viability and embryonic mortality of the American alligator in Florida. Ph.D. Dissertation. University of Florida, Gainesville Florida. 123 pp.

5.  Cardeilhac PT. 1987-1991. Diagnosis and treatment of infertility in captive alligators. Final Reports for the Aquaculture Market Development Aid Program. Vols. I-III. Florida Dept. of Agriculture, Tallahassee, FL 32399-0810.

6.  Millstein SR, RK Vander Meer, EM Schoenagle, PT Cardeilhac 1994. Dietary therapy for egg fertility in the American alligator: An evaluation by determining fatty acid profiles of egg yolk. Proceedings of the International Association for Aquatic Animal Medicine 25:10-15.

7.  Cardeilhac PT, ML Hoffingberg. 1985. Development of the Alligator. A Color Poster to Grossly Identify Developmental Stages of Normal Embryos Incubated at 88 F (31 C). Aquatic Animal Laboratory, College of Veterinary Medicine, Box 100126, HSC University of Florida, Gainesville FL 32610.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Paul T. Cardeilhac, DVM, PhD
University of Florida, College of Veterinary Medicine
Gainesville, FL

MAIN : Session VI : Moving Median Method
Powered By VIN