Spawning and non-breeding activity of adult giant bullfrogs (Pyxicephalus adspersus)

Abstract

Populations of the giant bullfrog (Pyxicephalus adspersus) have been poorly monitored due to the unpredictable appearance of this species aboveground. To better understand the activity of P. adspersus we quantified spawning by a population during five summers, and the activity of twenty adult frogs radio-tracked at the same site ca. twice weekly during the first three summers. In addition we examined animal activity, and population spawning in relation to meteorological variables, day of season, and moonlight. During the six-month summer period, males and females, respectively, spent 10±7 and 3±2 days at water, and 22±5 and 13±5 nights active. Greater proportions of radio-tracked animals moved overland, and/or foraged at night, around full moon, after heavy rainfall, when cooler, and less windy conditions prevailed. More animals were found at water, or on land during the day, and population spawning was more likely, earlier in summer, following heavier rainfall. Spawning occurred most frequently, in descending order, during December, January, and November, and was triggered by 40±16 mm rain in 24h. Spawning events lasted 2±1 days, but were prolonged around full moon. Numbers of males at spawning events varied between 30 and 500 males, and were positively correlated with the previous day's rainfall. Annually 6±1 spawning events occurred, and numbers of annual spawning events were positively correlated with total summer rainfall. Results of this study show that single counts of spawning adults will often result in gross underestimates of population size, and therefore, long-term adult counts are necessary to detect real population trends. Due to the unpredictable activity of adults, however, it may be more practical to monitor, within certain areas, the number of aquatic sites where breeding occurs, for improved conservation management of P. adspersus.

Keywords:

• Pyxicephalus adspersus
• Giant African Bullfrog
• breeding
• activity
• conservation
• rain

ACKNOWLEDGEMENTS

We sincerely thank Jimmy, Alice and Mark Yetman, and Johan Lötter for their patient and generous assistance with field work; Wendy Duncan for measuring local rainfall, reporting frog activity at the dams, and kindly assisting with field work; Roger Wood and Diepsloot Nature Reserve for providing additional rainfall data; the South African Weather Service for providing the bulk of the meteorological data; the Gauteng Department of Agriculture and Rural Development for issuing permits 1240 and 1296 to CAY; and Mike Perry, who introduced us to the site. The study was funded through the Endangered Wildlife Trust by Rand Merchant Bank, the Pretoria East branch of the South African Hunter's and Game Conservation Association, Arrow Bulk Marketing, Cellar Rats Wine Club, Bill Flynn and Diaz Films. We thank Michael Bates, Frank Lemckert, and three anonymous reviewers for valuable comments on previous drafts of this manuscript.