Seasonal and spatial changes in the larval fish fauna within a large temperate Australian estuary

A total of 66814 fish larvae, representing 37 families and 74 species, were collected in samples taken monthly between January 1986 and April 1987 from 13 sites located at frequent intervals throughout the large Swan Estuary in south-western Australia. The Gobiidae was the most abundant family, comprising 88.2% of the total number of larvae, followed by the Clupeidae (3.4%), Engraulididae (2.9%) and Blenniidae (1.0%). The most abundant species were Pseudogobius olorum (53.3%), Arenigobius bifrenatus (31.2%) and Engraulis australis (2.9%). Abundance of fish larvae in the lower, middle and upper regions of the estuary each reached a maximum between mid-spring and early summer, 2 to 4 mo before the attainment of maximum temperatures. Larvae of species such as Nematalosa vlaminghi and Apogon rueppellii were collected only between November and February, whereas those of others such as P. olorum, E. australis and Leptatherina wallacei were present over many months. The times and locations of capture of larvae have been related to the distribution and breeding periods of the adults of these species. The mean monthly number of species was far greater in the lower than upper estuary (14.7 vs 2.7), whereas the reverse was true for mean monthly concentration (42 vs 197 larvae per 100 m³). Classification, using the abundance of each of the 74 species recorded at the different sites, showed that the composition of the larval fish fauna in the lower, middle and upper estuary differed markedly from each other. Most larvae caught in the lower estuary belonged to marine species, whereas those in the upper estuary almost exclusively represented species that spawn within the estuary. The fact that the larvae of the 59 species of marine teleosts recorded during this study were restricted mainly to the lower estuary, and yet contributed only 6.2% to the total numbers for the whole estuary, helps to account for the relatively high species diversity in this region. The lack of penetration of many of these larvae beyond the first 12.5 km of the estuary presumably reflects the weak tidal effect in the wide basins of the middle estuary and saline regions of the tributary rivers. The larvae of the 13 teleosts that typically spawn within the estuary contributed 93.8% to the total numbers of larvae. Most of these estuarine-spawned larvae belong to teleosts that deposit demersal eggs and/or exhibit parental care (egg-guarding and oral and pouch-brooding), characteristics which would maximize their chances of retention within the estuary.     

Relationship between the habitat and diet of three species of atherinids and three species of gobies in a temperate Australian estuary

The distributions and diets of the six most abundant species of teleost in the shallows of a large south-western Australian estuary were examined from samples collected between March 1988 and February 1989. Fish were collected monthly by seine net from over bare sand and from within patchy and dense areas of the aquatic macrophyte Ruppia megacarpa. Their gut contents were compared with samples of the benthos and plankton collected from each of these three habitat types. The densities of the atherinids Leptatherina wallacei and Atherinosoma elongata and of the goby Favonigobius suppositus were greatest in dense R. megacarpa, whereas those of the atherinid L. presbyteroides and the goby F. lateralis were greatest over bare sand. The density of the goby Pseudogobius olorum was greater in patchy R. megacarpa than in the other two habitat types. The gut contents of each of the six species was dominated by crustaceans and/or polychaetes, with detritus also making a major contribution to the diet of P. olorum and F. lateralis. The relative proportions of prey items in the guts of fish in a particular habitat corresponded to the preys' relative occurrence in the environment. This indicates that the fish had been feeding predominantly in one particular habitat prior to capture. Within each habitat type, the six species partitioned the available food, the major components of the diet of each species being different. The gobiid species fed mainly on the benthos and the atherinids typically fed higher in the water column; A. elongata and P. olorum tended to be less selective as to where they foraged. There were no consistent differences in either the dietary breadth or the fullness of the guts of any species among habitat types.     

How colony growth affects forager intrusion between neighboring harvester ant colonies

Summary Colonies of the harvester ant, Pogonomyrmex barbatus, adjust the direction and length of foraging trails in response to the foraging behavior of their conspecific neighbors. In the absence of any interaction with its neighbor, a mature colony expands its foraging range at a rate of 0.85 ± 0.15 m per day. Exclusion experiments show that if a colony is prevented from using its foraging trails, the neighbors of that colony will enter its foraging range within 10 days. Exclusion experiments were performed with three age classes of colonies: young (1 year old), intermediate (3–4 years old), and old (5 years old or more). Colonies 3–4 years old are most likely to expand foraging ranges, and to retain newly-gained areas. To examine the relation of colony age (in years) and colony size (in numbers of workers), colonies of known age were excavated. Colonies increase greatly in size in years 3 and 4. Foraging area may be of greater current or prospective value for younger, smaller, quickly growing colonies than for older, larger ones of stable size. Correspondence to the second address     

Size and age compositions and reproductive biology of the nervous shark Carcharhinus cautus in a large subtropical embayment, including an analysis of growth during pre- and postnatal life

The lengths-at-age of individuals of the nervous shark Carcharhinus cautus in Shark Bay, Western Australia, have been determined and used to explore the types of situation when it might be advisable to shift from employing a von Bertalanffy equation to a more complex equation for describing the growth of this species and of elasmobranchs in general. The reproductive biology of C. cautus was also examined in order to construct curves for describing growth throughout life from conception as well as from parturition. The presence, in November and early December, of fresh bite marks on the sides of mature females and of a very high proportion of spent individuals among mature males indicate that C. cautus copulates in late October/early November. Ovulation and conception occur in late November/early December and parturition takes place approximately 11 months later. Since mature non-pregnant females contain vitellogenic ova for 12-13 months, i.e. from November or December to the following December, and mature pregnant females contain embryos for 11 months, i.e. from December to October, C. cautus has a biennial reproductive cycle. By parturition, the females and males of C. cautus had reached ~28% and 32% of their lengths at their maximum observed ages, respectively. The maximum recorded total lengths and ages of females and males of C. cautus were 133 cm and 16 years and 111 cm and 12 years, respectively. Females and males reached maturity at ~101 and ~91 cm, respectively, and at least 50% of females and males had become mature by the end of their sixth and fourth years of life after parturition, respectively. The three-parameter, von Bertalanffy growth curves provided reasonably good fits to the lengths-at-age of females and males of C. cautus during just postnatal life and throughout the whole of pre- and postnatal life. While the four-parameter, Schnute growth curve significantly improved the fit to these data for both females and males from conception and for females from parturition, it was recognised that the likelihood ratio test is very sensitive when, as in these cases, there are a large number of data points. A number of interrelated factors were thus taken into account when discussing circumstances when it might be appropriate to switch from using a von Bertalanffy growth curve to the more complex Schnute growth curve.