Observations of Breeding and Nesting Behaviors in Captive Varanus salvadorii Tim Trout Abstract Varanus salvadorii remains one of the enigmas of the monitor world. This paper deals primarily with the captive breeding of this species at the Denver Zoo. Successful reproduction occurred with one female on two separate occasions with two different males. However, behavioral data was collected during other non-successful breeding attempts as well. Parameters of successful hatching and rearing are described. Also, some unique nesting behaviors are discussed. Keywords: Squamata: Varanidae: Varanus salvadorii; reproduction, incubation. Introduction Unlike many of the other “giant” lizards of the world, very little is known about Varanus salvadorii’s natural history and ecology. In fact, not much more is known than where they come from and that they are considered the longest lizards in the world, reportedly attaining lengths of 4-5 meters (Bennett .998), although the largest substantiated measurement is 2.65 meters (Pianka et al. 2004). Because of the difficulties of studying these lizards in the wild, we must turn to captive observations to try to better understand them. Crocodile monitors are often displayed in zoos due to their large size, striking colors and tendency to sit in plain view of the public. For caretakers, they are also a prized species, because of their unique behaviors, intelligence and the challenge of propagation. It is only recently that captive propagation has occurred in V. salvadorii, with only four breedings documented so far (Horn & Visser .997 Pianka et al. 2004). Although it is likely that other undocumented breedings have occurred, this gives an idea as to the difficulty of captive propagation in this species. While progress is being made, captive breedings are still a rare occurrence, and those who do breed them often fail to replicate it. The Denver Zoo has had successes with V. salvadorii, producing offspring on two separate occasions from two different sires. This paper will deal with observations from these successful breedings—along with other unsuccessful egg-laying events—in attempts to try to contribute to the known reproductive behavior of this enigmatic species. Sex determination The first two crocodile monitors were obtained in September .999 from an importer and were set up in quarantine cages measuring 3m x ..5m x ..5m (L x W x H). These animals were 500g and 700g respectively and were estimated at around 2 years of age. Another animal was obtained in November .999, weighing .600g and estimated at 3-5 years of age. Based on appearance, growth rates, and behavior it was believed there was a 2.. sex ratio. In March 2000, we wanted to see if we could confirm this belief. Various methods of sexing monitor lizards have been described (Bennett .998, Horn & Visser .997). Radiographs and ultra-sounds were performed to try to determine the sexes of the animals. While x-rays proved inconclusive, the ultra-sounds clearly showed two males Tim Trout Fig. 1. Ultrasound image of female V. salvadorii. Sex SVL (cm) Total length (cm) Weight (g) Male . 45 .53 .900 Male 2 Female 49 43 .69 .35 2400 .430 Tab. 1. Size at which animals were accurately sexed using ultrasound. and one female. See Fig. . for female identification. The sizes of animals at the time of sexing are listed in Tab. .. Husbandry Care for these monitors is relatively straight-forward for a large active lizard. A large walk-in enclosure is preferable. An enclosure size of 2m x 2m x 2m (L x W x H) should be considered the minimum for an adult. Ambient temperatures anywhere from the upper 20’s°C to 30°C are adequate, but a basking spot of at least 40°C should always be provided during daylight hours (up to 65°C has been used successfully). As long as they are able to move to cooler areas, I believe the hotter the better for basking. Another critical element to successful husbandry is access to UV-B light. At the Denver Zoo, we use Weisslite brand 300W spotlights, which provide significant UV-B at several feet away and also provide intense white light and heat. In addition, over our V. salvadorii exhibit we have added a UV-transmiscient panel to our roof. While the animals definitely respond to and follow the light from this panel throughout the day, the artificial sources of UV are more than adequate for captive maintenance. Feeding is straight-forward. From hatchling to adult, they readily take pre-killed rodents. Babies will sometimes need the rodent eviscerated to stimulate feeding. For problematic imports, nothing works better than live rat pups (.5-20g size). For babies, I offer 20% of lizard’s body weight per week spread over three feedings every week. For growing juveniles, I offer 20% of body weight weekly over 2 feedings. Once adult, females get .0% of body weight in one feeding per week. Adult males get 5-7% of body weight every .-2 weeks. Females being conditioned for breeding seem to do better with a little extra body weight; whereas males seem to breed better if kept on the lean side. Moisture is another important consideration for V. salvadorii. Ambient humidity Breeding and Nesting Behaviors in Captive V. salvadorii Fig. 2. Naturalistic habitat at the Denver Zoo. Note the plump — and possibly gravid — V. salvadorii at the rear of the exhibit. should preferably be above 60%. In addition to this, regularly misting the animal directly is critical. While I have never experienced any pathology from lack of contact moisture, the skin takes on a dry and unattractive appearance. Providing a large water container for them to soak in will not likely help skin condition, as I have rarely observed them to soak even though a large enough water receptacle has always been available. There is also the possibility that newly imported animals will only drink from mist, although they will soon start recognizing a small water bowl. Interestingly, I have observed V. salvadorii not drinking from a bowl when it is too large (i.e.—60cm diameter, 20cm deep); however, when a smaller one is introduced (i.e.—20cm diameter, 8cm deep) they drink heavily. Breeding In June 200., a male and female were housed in wire holding cages next to each other, with a visual barrier present. It was observed that for several weeks there was substrate on the ground outside of the male’s cage daily. I suspected he may be restless as a consequence of the female ovulating and decided to put them together. On July 2, 200. both were placed together in a large planted exhibit in the Denver Zoo’s Tropical Discovery building. Initially there was some aggression resulting in bite wounds. Over the next week, they were separated overnight, as keepers were not in the building to separate them if they fought. Eventually the fighting discontinued. In early August, the female began digging holes throughout the exhibit, which has soil substrate up to approximately 75cm deep. At this point it was assumed she was gravid—due to her swollen appearance— and the male removed for fear he may eat the eggs. She continued digging in numerous locations until October .6, 200., when she laid a clutch of 7 eggs (an 8th egg was laid the next day). The nest hole was 30cm deep and 20cm in diameter. The eggs were immediately removed for artificial incubation. The female however, continued with the nesting behavior. Once finished, it was no longer apparent where the eggs had been laid. She not only filled the nest to the point that the soil where the eggs were laid appeared compacted, but she also used her claws to rake the remainder of the soil Tim Trout in the enclosure. It is hypothesized by the author that this was done so predators would not recognize the actual nest site and prey on the eggs. After egg-laying the female appeared to defecate exclusively on the nest site. This lasted for over three months. It is hypothesized that this was done to scent camouflage the nest, as she spent significant time away from the nest in the exhibit. Over successive breedings (5 clutches of eggs recovered and 2 others suspected laid and immediately eaten), the above description was generally applicable. There are minor differences from time to time, but this should be considered a guideline. One difference involved the actual breeding introduction. Following the above mentioned breeding, the female has been kept on display in the heavily planted exhibit and males have been introduced for breeding purposes only. It seems that nothing particular is necessary to stimulate copulation other than the introduction of a male into the female’s territory. Every occasion a male has been introduced, copulation has occurred which usually—though not always—leads to egg-development. Tab. 2 includes dates of egg deposition. Note that most egg depositions occur within a 3-4 month period, with one exception (June 22, 2002), which resulted in a successful hatching. Tab. 2. Dates and eggs laid Tab. 3. Measurements of known fertile at the Denver Zoo. eggs. The bottom row indicates the average of the individual measure- Month . of eggs ments above. Oct .6, 200. 7 Weight(g) Length(cm) Width(cm) June 22, 2002 7 88.6 9.0 4.2 Jan 29, 2003 9 88.. 8.8 4.2 Oct .3, 2003 6 90.0 8.8 4.2 Nov ., 2004 4 83.6 8.5 4.3 88.5 8.8 4.2 88.6 8.2 4.3 89.6 9.2 4.2 89.2 8.7 4.3 85.3 8.7 4.2 85.5 8.5 4.. 87.7 8.7 4.2 When this particular female gets close to actually laying her eggs, she becomes aggressive towards males: chasing, mounting and even biting them, causing significant injury. This may not necessarily be the case with all V. salvadorii, as others have reported males actually assisting in nesting behaviors (Meier, pers. comm.). Incubation Incubation at the Denver Zoo has always been challenging, particularly for delicate eggs. There are two significant issues that we must overcome. The first is the dryness of the climate, which is often not more than 20% humidity. While a scenario of near .00% humidity can be created in an enclosed box, they seem to dry out much more quickly Breeding and Nesting Behaviors in Captive V. salvadorii Tab. 4. Egg size comparison between Schmicking & Horn and current paper. Tab. 4. Egg size comparison between Schmicking & Horn and current paper. Average mass (g) Average length (cm) Average width (cm) Schmicking & Horn 60.8 8.8 4.0 Trout 87.7 8.7 4.2 % difference 3.% .% 5% and we have found that a moister incubation medium is required. The second—and more difficult—issue in Denver is the altitude. Being around .600m above sea level means less oxygen in the air, and this appears to affect hatching. Interestingly, this has also been the case with Denver Zoo’s bird incubation parameters (Haeffner, pers. comm.). On numerous occasions with different types of reptiles, we will encounter seemingly normal embryos that go full-term and – for no obvious reason – just do not hatch. One hypothesis is that the availability of less oxygen means the egg must operate at a higher metabolism, and thus runs out of yolk before it is ready to hatch; thus, dying before pipping the egg. Through trial and error, we seem to be able to overcome the hatching problem by using slightly cooler temperatures (.-2 °C). By incubating cooler, the embryo’s metabolism will be relatively slower as well, allowing them to develop full term. Successful incubation has been achieved at a ratio of .:. by mass of water vermiculite to water at 28-29 °C. Hatching occurs at an average of 20. days (with a range of .89 – 2.. days). A critical element to V. salvadorii eggs incubation is how the eggs develop. Through approximately the first two-thirds of incubation the eggs will swell, increasing both in size and mass. During the final third of incubation, the eggs shrink so that at hatching they are the same size and mass as when laid. This is due to water absorption and expulsion. Because of this, the eggs are kept in a box without ventilation for the first 2/3 of incubation, and boxes are weighed and water replenished as needed to maintain the .:. vermiculite:water ratio. During the final third of incubation, no water is replenished to the incubation medium and ventilation holes are added. The amount of ventilation can be determined by the amount of condensation observed in the egg container. The goal should be no observable condensation. The idea is that by equalizing the air inside with that outside, the developing eggs are able to expel all the water; thus preventing them from drowning inside the egg prior to hatching (this has occurred in V. salvadorii and V. komodoensis at the Denver Zoo). The eggs will generate significant heat through metabolism near the end of incubation. The temperature differential between inside and outside the egg box will cause condensation, thus preventing the nec- Tab. 5. Hatchling data for 2 separate clutches of 5 total hatches at Denver Zoo. Hatch date Mass (g) SVL (cm) Total length (cm) Jan 9, 03 68.0 .6.0 47.5 Aug 6, 03 45.9 .5.5 45.0 Aug .., 03 42.4 .4.0 4..0 Aug 25, 03 66.8 .6.0 49.5 Aug 28, 03 69.4 .6.5 50.6 AVERAGE 58.5 .5.6 46.7 Tim Trout Fig. 3. A hatchling V. salvadorii at the Denver Zoo. Photo by RICK HAEFFNER. Fig. 4. V. salvadorii hatching. Photo by ALEKSEI SAUNDERS. essary expulsion of the moisture absorbed during the first part of incubation. As a sidebar, another incubation method was tested using what is referred to as the “hygroscopic method” (Seward 2002). The theory is to provide an environment of Breeding and Nesting Behaviors in Captive V. salvadorii .00% humidity, without the eggs actually having contact with water in the substrate. This method was successful in hatching V. salvadorii; however, based on observations of eggs in both this and the traditional method outlined above, it seems the hygroscopic method is not ideal for hatching this species. The contact moisture of incubating the eggs in moist vermiculite appeared to have a distinct benefit to the eggs. In Tab. 3 egg weights and sizes for ten known fertile eggs are listed. These were compared to the measurement listed by Schmicking & Horn .997. Interestingly, while the lengths and widths are similar, the mass is significantly larger on the eggs I have seen. The differences are noted in Tab. 4. Hatchling measurements are listed in Tab. 5. Discussion While the observations detailed above are only representative of a single female V. salvadorii’s breeding habits, hopefully they can be generalized to others of this species. Because it is quite difficult to study them in the wild (for various reasons), much of their habits and biology will have to be determined in a captive setting. Observations of the Denver Zoo’s V. salvadorii are that they do not necessarily have a specific breeding season in captivity, outdoor environments are not necessary for propagation, and the females display nesting-guarding behavior for several months after egglaying. With every observation and successful breeding, we can come closer to understanding these fascinating and unique monitors. In the future, breedings will become commonplace and the species will gain a stronghold in captive populations. Because of the –generally – more docile temperament of captive offspring, they are much easier and safer to work around. Hopefully, this will increase the number of institutions willing to house and exhibit them so that many can marvel at what is truly one of the planet’s most unique reptiles. Zusammenfassung Varanus salvadorii bleibt eine der rätselhaftesten Arten in der Welt der Warane. Der Beitrag beschäftigt sich hauptsächlich mit der Nachzucht dieser Art im Zoo zu Denver / USA. Eine erfolgreiche Nachzucht gelang mit einem weiblichen Tier mit zwei verschiedenen Männchen in zwei verschiedenen Nachzuchten. Eine Reihe von Beobachtungen zum Verhalten dieser Art bei nicht erfolgreichen Reproduktionsversuchen werden ebenfalls mitgeteilt. Parameter zum erfolgreichen Schlupf und zur Aufzucht der Jungtiere werden angegeben. Ferner werden einige ungewöhnliche Beobachtungen zum Nistverhalten beschrieben. Schlüsselwörter: Squamata: Varanidae: Varanus salvadorii; Fortpflanzung, Inkubation. References Bennett, D. (.998): Monitor Lizards.—Edition Chimaira, Frankfurt am Main, 258 pp. Horn, H.-G. & G. J. Visser (.997): Review of reproduction of Monitor lizards in captivity II. In ternational Zoo Yearbook 35: 227-246. Pianka, E. R., King, D.R. & R. A. King (2004): Varanoid Lizards of the World. Indiana Univer sity Press, Bloomington & Indianapolis. 237 pp. Schmicking, T. & H.-G Horn (.997): Beobachtungen bei der Pflege und Nachzucht des Papua warans, Varanus salvadorii. – Herpetofauna, Weinstadt, .9(.06): .4-23. Seward, M. (2002): Dr. Seward’s Gila Monster Propagation. – Natural Selections Publishing. 53-60. Author: Tim Trout, 10850 W. 38th Pl., Wheat Ridge, CO 80033, U.S.A.; e-mail: TTrout@Denverzoo. org.