We investigated whether breeding frequency and breeding success of southern Buller’s mollymawks (Thalassarche bulleri bulleri) were influenced by breeding experience and pair bond duration, using data from annual checks at 3 study colonies at The Snares from 1992 to 2001. Most pairs bred annually irrespective of the experience and length of the pair bond, although the proportions that did so varied with pair type. Thus, breeding frequency (% breeding in consecutive years) was lowest among pairs of 1st-time breeders (77%). Breeding frequency of those pairs after their 2nd attempt (89%) became similar to that of established pairs together for at least 1 previous breeding attempt (88%), or newly formed pairs in which one or both birds had previous breeding experience (91%). Overall breeding success was 71% and, in established pairs, breeding failure (loss of egg or chick) was associated with reduced breeding frequency (83% compared to 91% when successful). Lowest breeding success (58%) was associated with the attempts of 1st-time breeders. Performance of these pairs improved until the 3rd attempt (81%), when it became similar to that of established pairs (73%) and newly formed pairs in which one or both birds had previous breeding experience (77%). Divorce was rare (1.1-3.5% annually). First-time and former breeders mated more frequently with birds of similar status (85% and 58% respectively) than expected assuming random pairings. When changing partner, as a result of divorce or death, the average interval before breeding again was 2.1 years for males and 2.6 years for females, and so, on average, each change of partner resulted in the loss of 1 breeding attempt. Thus, the time taken to obtain a new partner has a lifetime reproductive cost.
We studied the breeding biology of a colony of Caspian terns (Sterna caspia) near Invercargill, New Zealand, during 1992 and 1993. The mean clutch size did not differ between years and averaged 2.04. Measurements of 147 eggs averaged 64.5 x 44.6 mm; there was no difference in size of A-eggs (1st-laid in a clutch) and B-eggs (2nd-laid) in either year, but the few C-eggs laid were significantly smaller. The incubation period averaged 27.2 days (range 26-29 days); some earlier published values of 20-22 days appear to be in error. In 1992, growth rates of A-chicks were significantly higher than those of B-chicks. Growth rates of A-chicks were significantly higher in 1992 than in 1993. Fledging occurred at 33-39 days at an estimated average mass of 527 g in 1992 and 501 g in 1993. Minimum productivity was 1.04 and 0.62 chicks fledged per pair in 1992 and 1993, respectively. Weather during the period of chick growth was much wetter and windier in 1993 and we suggest that this reduced the ability of parents to feed chicks. Investigator disturbance, which has been implicated in lower reproductive success in some studies of Caspian terns, did not appear to have a major impact in our study. We believe this was partly because the birds were habituated to our activities and partly because of our methodology.
The occurrence of musk ducks (Biziura) as fossils in New Zealand is reviewed and updated. Twenty-four bones from at least 7 individuals, and 67 elements from a single skeleton are known. Morphological differences between the fossils and the extant Australian B. lobata support continued distinction of the New Zealand form as the separate species B. delautouri.
The night-time activity of grey-faced petrels (Pterodroma macroptera gouldi) was measured at a colony on Tiritiri Matangi Island between 27 April and 10 December 1998. Considerable seasonal variation was observed (0 to >120 birds/night). A decline in numbers of birds at the colony in early June was likely resulted from the departure of both breeding (pre-laying stage) and non-breeding birds. Another decline at the end of September was most likely a consequence of the departure of non-breeding birds only. In general, as the season progressed there were fewer petrels per night, and they arrived later. The number of birds returning to the colony increased with increasing wind speeds and birds arrived earlier when winds were stronger. High wind speeds facilitate movement between breeding and foraging grounds for this pelagic species. A sampling period of 1 h from the arrival of the 1st bird provides sufficient information to discern definite patterns in numbers throughout the year.
Observations of eastern rosellas (Platycercus eximius) in the Wellington region were recorded during February 1994 to January 1997 by members of the Ornithological Society of New Zealand. Sixty-one percent of the 1227 sightings were made in 3 localities, Paraparaumu-Waikanae, Upper Hutt, and central Wellington. Most rosella sightings were in pastoral farmland (47%) and urban habitats, including parks, and golf courses (41%). Of the 400 sightings for which flock size was given, the mean was 3.6 rosellas, and flock size did not change significantly with season. Rosellas were seen feeding (n = 54) on buds and shoots (19%), flowers (5%), fruit (15%), and seeds (61%), with feeding on introduced plants making up 80% of the feeding observations.