Search

The relationship between sediment type and the distribution of the commercial prawn species of the western Gulf of Carpentaria was examined. The distribution of sediments was described on the basis of the mud content while the spatial distributions of the adult populations of all the commercial species were described from the results of trawl surveys. The main species caught were the tiger prawns Penaeus esculentus and P. semisulcatus, the endeavour prawns Metapenaeus endeavouri and M. ensis, and king prawns P. latisulcatus and P. longistylus, the banana prawn P. merguiensis, and the coral prawn Solenocera australiana. The individual species were caught in varying depth ranges and, in order to assess the influence of sediment type on the spatial distributions, a stepwise multiple regression analysis was carried out in which variation due to depth was considered before that due to percentage mud. Although depth generally accounted for most of the variation in catch-per-unit-effort (CPUE), percentage mud was also found to be a significant factor for all species except P. Merguiensis. Three species, P. semisulcatus, M. ensis and S. australiana, showed a preference for sediments with a high mud content while the abundances of P. esculentus, M. endeavouri and P. latisulcatus were each negatively correlated with percentage mud. P. longistylus showed a strong depth-mud interaction, being largely found on sediments of 40-60% mud in depths of 40-50 m. P. merguiensis was found in depths 75% mud) whereas those of P. esculentus were associated with coarser sediments (50-75% mud).

The food of seven species of penaeid prawns from the Gulf of Carpentaria consists predominantly of Foraminifera, small molluscs, crustaceans and polychaetes. Measurements of organic and inorganic carbon, organic nitrogen and bacterial biomass were made. Foregut contents of adult prawns contained between 72 and 223 mg organic carbon/g dry wt. Protein constituted between 43 and 64% of the organic matter. Approximate assimilation efficiencies of food in prawns caught in the gulf, determined for four species, varied from 48 to 77% of organic carbon and from 42 to 77% of organic nitrogen. The food of juvenile Penaeus merguiensis was examined for two growing seasons. In the 1976-1977 season the foregut contents contained a mean of 41 mg organic nitrogen /g dry wt and 181 mg organic carbon /g dry wt. In the 1977-1978 season, significantly lower proportions of organic nitrogen and carbon were eaten, viz, 21 mg organic nitrogenlg dry wt and 101 mg organic carbon /g dry wt. Improved assay procedures for muramic acid have shown that bacteria are less important in the food of prawns than previously reported. Bacteria constituted less than 2% of the organic matter in the adults of all species, but in many juvenile P. merguiensis bacteria were more important, constituting up to 14% of organic matter.

From 1981 to 1984, aerial surveys along the north-east Australian coast between Cairns in Queensland and Milingimbi in Arnhem Land recorded more than 250 000 migratory wading birds in December and February and up to 60 000 in July and early September, largely in the south-east Gulf of Carpentaria. The pattern of dispersion of waders was similar in both summer and winter. More than 85% in both seasons were found along muddy coastline fringed by mangroves. Waders congregated near the mouths of the rivers with the largest catchments.

Starch gel electrophoresis of enzymes and polyacrylamide gel electrophoresis of muscle proteins were used to study genetic variation in 589 barramundi from three widely separated regions in northern Australia. Presumed genetic variation was observed at 16 loci (out of 46 screened). However, only 11 of these were polymorphic at the p0.99 level. The average heterozygosity in this species was 0.032. In nearly all cases, genotypic proportions agreed closely with Hardy-Weinberg expectations. Repeated sampling in each region revealed little or no change in allele frequencies over a period of several months. Similarly, comparisons of allele frequencies for fish from marine and from freshwater localities in the south-eastem Gulf of Carpentaria failed to reveal significant genetic differences between habitats. Between-region heterogeneity Χ2 tests indicated substantial genetic differentiation at 10 of the 11I polymorphic loci. These data indicate the existence of at least three distinct stocks or subpopulations of barramundi in Australia.

Parasites in banana prawns, P. merguiensis, from the Gulf of Carpentaria were surveyed. Only the lecanicephalid cestode Polypocephalus sp. seemed useful as a biological marker for banana prawns. Loads of Polypocephalus sp. were low and constant at salinities from 14 to 34. At higher salinities (38), both intensity and prevalence of infection rose markedly from 0.5 Polypocephalus sp. per prawn and 27% to 6.25 and 95%, respectively. The frequency distribution of Polypocephalus sp. was adequately described by a non-truncated form of the negative binomial distribution (n = 943, p = 35.4, k = 0.43) suggesting there were no parasite-induced mortalities. Loads of Polypocephalus sp. in prawns in the commercial catch from Albatross Bay were most similar to loads in prawns from the Mission River, which empties into Albatross Bay. Polypocephalus sp. loads in Penaeus merguiensis formed two other natural groupings, one in the south-east corner of the Gulf and one in the Cape Keerweer-Mitchell River area. There was no parasitological evidence for extensive migrations of adolescent P. merguiensis away from their nursery estuaries.

A trawl survey of the south-eastern Gulf of Carpentaria carried out at the beginning of a commercial prawn fishery in the 1960s provided data on the demersal fish of an unexploited tropical fish community. The extent of temporal variation in this community was investigated using abundance data on 359 taxa of fish and cephalopods over 13 months in 1963-1964. Seasonal changes in water temperature and salinity were large, particularly in nearshore waters. Catch rates and species richness in nearshore waters were highest during summer, and in offshore waters during autumn or winter. Shannon diversity was variable, with no clear seasonal component. Temporal changes in community composition resulted in large changes between different seasons in the structure of site groups derived by classification; temporal effects within seasons were also found. The numerically dominant species were the leatherjackets Paramonacanthus spp., the ponyfish Equulites leuciscus, the tripodfish Tripodichthys blochii, and the saury Saurida undosquamis. Squid, the ponyfish Leiognathus sp. nr blochi, the butterfly-bream Nemipterus tolu and the grunter Pomadasys maculatus were seasonally abundant. Although species were restricted in their depth range, discrete communities that maintained their identity in different seasons were apparently absent. The relative abundance of many species varied substantially wirh season and probably also over a longer period. Small changes in local abundance were often associated with movement to deeper water at times of high nearshore water temperature (summer) or of reduced nearshore salinity (autumn). Large seasonal changes occurred in the local abundance of estuarine and semidiadromous species. The community shared many species with the demersal fish community of the Gulf of Thailand. It is suggested that similar structural changes in the fish community of the Gulf of Carpentaria may occur in response to intensive fishing. Further research on the effects of demersal fishing and on the interactions of the demersal fish community with commercially important crustaceans is necessary.

Fish communities in the south-eastern Gulf of Carpentaria were surveyed before the development of a demersal trawl fishery for prawns. Distribution data for fish and cephalopods were examined to determine community composition and species-distribution patterns in an unexploited shallow-water tropical fish community. Common species were mostly Indo-west Pacific in affinity, and included many species important in demersal trawl fisheries. Three clusters of site groups were present, comprising site groups from nearshore (~2-5 m), shallow offshore, and deep offshore (~15-35 m) areas, respectively. Site groups did not correspond to clearly defined communities. Species overlaps were frequent. Limits to the distribution of common species seemed to be related more to physical factors (bottom depth, salinity and sediment type) than to other biotic factors (distribution of species). The overall picture was of a rich assemblage of fish species in an environment characterized by gradients rather than by sharp changes.

Caranx bucculentus is an abundant predator in the commercial prawn grounds of Albatross Bay in the Gulf of Carpentaria. The stomachs of samples taken from Albatross Bay contained mainly teleosts (48%) and crustaceans (46%), with commercially important prawns contributing 15% by dry weight to the total diet. The proportion of teleosts, penaeids, brachyurans, 'other crustaceans' and molluscs in the diet changes as the fish grow and their diet shifts from mainly crustaceans to mainly teleosts. Although the percentage of teleosts, penaeids, brachyurans and 'other crustaceans' eaten varied with the time of sampling, no seasonal variation was noted. Only the numbers of penaeids and 'other crustaceans' eaten varied significantly over 24 h. Caranx bucculentus less than 200 mm SL ate most penaeid prey during the day and primarily took small, commercially unimportant species. In contrast, fishes more than 275 mm long took larger, commercially important species, and mainly at night. The abundance of large C. bucculentus and the high incidence of commercial penaeids in their diet indicate that this species may have a considerable impact on the prawn resources of Albatross Bay.

To provide the information necessary for optimizing sampling strategies, the effects of tidal, diurnal, lunar and seasonal cycles on the catchability of juvenile and adolescent banana prawns, P. merguiensis, were examined in the Norman River and in the inshore coastal zone of the Gulf of Carpentaria. A major part of the large variability in catches in both environments was attributable to marked tidal periodicity in catchability. In the river, large catches of juvenile P. merguiensis were taken only at or near the time of low water during both the neap and spring tide periods. Smaller prawns (< 12 mm carapace length) tended to be even more restricted to the time of low water than were larger prawns. This phase shift in catchability of larger prawns was particularly evident in the offshore sampling where large catches of adolescent prawns were made only at high water. Both the number and size of prawns caught also varied considerably with water depth. The reasons for the marked differences in catchability throughout a tidal cycle are not clear, but it is obvious that sampling can be restricted in time, thus providing much more consistent data with a minimum of effort.

The histological events associated with gonad maturation in male and female L. calcarifer are described. Females are normally derived from post-spawning males. A histological description of the process of sex inversion is presented. The lengths of males at first maturity in the Northern Territory and south-eastern Gulf of Carpentaria were 600 and 550 mm, respectively. Most fish were changing sex at lengths of 900 and 850 mm, respectively in these areas. These size differences were due to a slower growth rate of barramundi in the Gulf of Carpentaria, both processes being related to age rather than to size.

The reproductive cycle of female P. esculentus was investigated in the Groote Eylandt region of the Gulf of Carpentaria from August 1983 to March 1985. The minimum size at maturity for P. esculentus was 25 mm carapace length (CL), and 50% of the population were mature at 32 mm CL. The proportion of females which had mated increased sharply above 28 mm CL to a maximum of 80% inseminated in the size range 32-50 mm CL. An index of population egg production, calculated from female abundance, the proportion of females spawning and fecundity with size, was used as an indicator of reproductive output. Egg production tended to be spread throughout the year, but with eggs being produced most consistently in late winter and early spring. Spawning occurred in a limited area within the study area. A comparison of P. esculentus and P. semisulcatus showed that P. esculentus matures at a smaller size (50% at 32 mm CL) than P. semisulcatus (50% at 39 mm CL), fecundity is lower, spawning is nearer inshore and egg production is less strongly seasonal.

Data from three current-meter moorings in the south-east corner of the Gulf of Carpentaria indicate that barotropic diurnal tidal currents are predominant. The low-passed currents are also barotropic but there is not a good correlation between the currents at the three moorings or with the wind recorded at Mornington Island. This may be due to the existence of topographic gyres in the residual currents, or the winds recorded at Mornington Island not being representative of those at the mooring sites.

Aerial surveys of coastal waters (27 216 km2) in the western Gulf of Carpentaria during the dry season of 1984 and wet season of 1985 indicated no major seasonal changes in distribution and relative abundance of dugongs. Minimum population size in the dry season was estimated as 16 816 � 2946 (standard error), with a relative density of 0.62 k 0.11 km-2, and that for the wet season 16 846 + 3257, with a relative density of 0.62 � 0.12 km-2. The estimates exclude 5% of observations which could have been either dugongs or Irrawaddy dolphins, and were corrected for submerged dugongs and those missed on the surface. Dugongs were unevenly distributed over the study area, and neither mean group size nor proportion of calves varied between seasons. Dugong abundance was correlated with area of available seagrass. The catch rate of dugongs by Aboriginal people off Numbulwar decreased six-fold between the 1960s and 1985 (60 to 10 p.a.), possibly due to excessive hunting.

Cyclone Kathy on 23 March 1984 stranded many green turtles, Chelonia mydas, on the coast adjacent to the Sir Edward Pellew Is. It is estimated that over 1000 were thrown up by the storm surge and that over 500 were left stranded by the receding waters, including migrants from the Raine I. rookery. These turtles, which were predominately large females, had been feeding close inshore on seagrass. Of the mature females sampled, 62% were preparing to breed in the next breeding season but none had bred in the previous season. The flatback turtle, Chelonia depressa, the principal species breeding on the Sir Edward Pellew Is, was not involved in the stranding. Hawksbill turtles, Eretmochelys imbricata, and olive ridley turtles, Lepidochelys olivacea, which occur in the deeper offshore waters were also not stranded. Some of the rescued turtles were subsequently recaptured nesting at Raine I.

A non-linear barotropic model was used to evaluate the tidal regime in the Gulf of Carpentaria. The model was forced by open boundary conditions specified on a line joining Wessel Islands to False Cape and a volume flow through Torres Strait estimated from tidal constants on each side of the Strait. The model gives results in agreement with the available observations and in particular predicts mixed tides in the northern half of the Gulf and diurnal tides in the south-east corner of the Gulf. The diurnal tide consists of a Kelvin wave entering the Gulf in the north-west and propagating clockwise around the Gulf with one amphidromic point. The higher frequencies of the semidiurnal tides allow the generation of a first-mqde Poincare wave and the trapping of energy in the northern half of the Gulf. Amphidromes near Mornington Island and Groote Eylandt are also predicted, as is a region of low amplitude and rapid phase variations in the centre of the Gulf.

A stock assessment of P. merguiensis in the Gulf of Carpentaria was made by means of a yield per recruit analysis based on studies of migration, growth and mortality. Data were collected both from the commercial fishery and tag recapture experiments. No significant movement of tagged prawns out of the fishing grounds occurred during the fishing season. Estimates of the von Bertalanffy growth parameters (L∞ = 38.0 mm carapace length, K = 0.08 week-1) were obtained from the change in monthly size distribution of commercial catches. The natural mortality coefficient (M) was 0.05 week-1 while the total mortality coefficient (Z) estimates for fishing seasons during 1974-1976, ranged from 0.22 week-1 to 0.35 week-1. The corresponding estimates of the optimal size at first capture ranged from 30.6 to 32.6 mm carapace length. Despite the high rates of exploitation (E = 0.78-0.86), there was no evidence to suggest that recruitment had been adversely affected.

Growth data were obtained for the two tiger prawn species P. esculentus and P. semisulcatus from a tagging experiment carried out in February 1981 in waters adjacent to Groote Eylandt in the western Gulf of Carpentaria. A von Bertalanffy growth curve was fitted to these data and least squares estimates of the parameters L∞ and K and joint 95% confidence regions were calculated for males and females of both species. Tests on the residuals from the fitted curve were carried out to check the adequacy of the fit of the von Bertalanffy model, and differences in the parameters L∞ and K between sexes and species were examined. An alternative von Bertalanffy-type model was used to test the consistency between the estimated L∞ values and the carapace leilgths of the largest individuals observed in samples from the populations. For males of each species, the fit of the von Bertalanffy model was satisfactory, and the estimates of L∞ were consistent with the largest observed carapace lengths in catch samples. For females, however, the fit of the von Bertalanffy model was not entirely satisfactory, especially for P. semisulcatus, for which additionally the estimated L∞ was much higher than the largest carapace length in catch samples. Possible reasons for this lack of fit are discussed. Estimates of L∞ all differed significantly, with the exception of P. esculentus males and P. semisulcatus males. No significant difference was found in estimates of K for P. esculentus males and females, and only a marginally significant difference in K between P. esculentus males and females and P. semisulcatus males. Presence of a bopyrid parasite in P. semisulcatus did not affect growth.

Trawl surveys in the north-western Gulf of Carpentaria were carried out each lunar month from August 1983 to March 1985 to assess the temporal and spatial distribution and abundance of P. esculentus and P. semisulcatus. The information obtained was then compared with that from fishermen's logbooks. Water temperature and salinity were monitored during the study and their possible influence on the distributions has been inferred. The distributions of juveniles of less than 20 mm carapace length indicated that, for both tiger prawn species, the main nursery areas in the region were in Blue Mud Bay and in the bays along the northern coast of Groote Eylandt. Although the two species shared the same nursery areas, the juveniles were concentrated in different parts of Blue Mud Bay and were most abundant at different times. Catches of juvenile P. esculentus increased substantially in October and peaked in November, whereas catches of juvenile P. semisulcatus increased in November and peaked in January. Catches of both species showed a secondary peak in March 1984, coincident with the heaviest monthly rainfall of the summer monsoon season. The distribution of larger prawns showed spatial separation of the two species in the offshore fishery. The monthly pattern of the catch-per-unit-effort (CPUE) of the two tiger prawn species combined was similar to that obtained from fishermen's logbooks; the logbook data are therefore generally reliable. Monthly CPUE from the trawl surveys for the individual species showed distinct peaks in late summer, which were apparently related to recruitment of small prawns into the fishery. CPUE for P. semisulcatus peaked in February and April (juvenile abundance had peaked in January and March); CPUE for P. esculentus peaked in January and May (juvenile abundance had peaked in November and March). However, a distinct CPUE peak in spring (August/September) for P. semisulcatus could not be related to a previous peak in juvenile abundance; this was presumably a result of an increase in catchability. Although there was evidence linking changes in the catchability of P. semisulcatus to changes in water temperature, a similar link was not as evident for P. esculentus.

The Carpentaria Basin is an epeirogenic intracratonic downwarping which formed during the Mesozoic. It is the northernmost of three main Jurassic-Cretaceous downwarps forming the Great Australian Artesian Basin. The basin is filled with clastic sediments, predominantly derived from Precambrian ridges on the flanks of the basin, which thicken gradually from the basin margins towards the depocentre.The Carpentaria Basin is analogous to the Eromanga Basin immediately to the south, and several rock units were deposited contemporaneously in both basins so that they can be mapped in the subsurface from one basin to the next. There is no change in sedimentary style or lithology.The character of the Base Mesozoic Unconformity, over which the Carpentaria Basin was deposited, changes abruptly from east to west. East of a point approximately coincident with the Queensland-Northern Territory border the unconformity surface is flat and featureless but to the west it becomes very rugged and irregular. This rugged horizon is believed to reflect either ancient karst topography or extensive reefal build-up equivalent to dolomite intersected in Burketown 1. Based on correlations with the McArthur and Georgina Basins it is probably Cambrian or Proterozoic in age. These rocks are known elsewhere to be porous and permeable.The unconformity shelves to the west towards Groote Island and the Northern Territory mainland and sediments of the Carpentaria Basin onlap a shallowing ancient landscape. The western boundary of the basin believed to be prospective for hydrocarbons is taken as the limit of deposition of the Cretaceous Gilbert River Formation.Lithologies and the existence of aquifers in the onshore Carpentaria Basin indicate that suitable reservoirs are present. Sandstones in the Cretaceous Gilbert River Formation, the Jurassic Eulo Queen Group and the Hutton Sandstone have the best reservoir potential. The Gilbert River Formation is the primary objective because it was extensively deposited throughout the basin, whereas Jurassic sandstones are restricted to topographic lows.The Gilbert River Formation is sourced and sealed by shales of the Wallumbilla Formation. Onshore this formation is too shallow to be mature but seismic indicates that offshore it is sufficiently deeply buried to be producing and expelling hydrocarbons.The hydrocarbon prospects of the Carpentaria Basin have in the past been considered low because the Mesozoic sequence onshore is thin and even offshore the total thickness of Mesozic and Cainozoic sediments is only about 1760 m at its maximum. This judgement ignored the fact that, although no economic hydrocarbon discoveries have been recorded from the Carpentaria Basin and/or the underlying Palaeozoic Basins, several shows have been reported. Shows reported in onshore wells suggest hydrocarbon generation within the basin. Geochemical and palynological studies carried out on samples from onshore wells and bores indicate the basal Mesozoic sequence is within geochemical Zone II, where the actual generation of hydrocarbons occurs but where significant mobilization and effective oil expulsion do not take place. These data suggest the deeper offshore portion of the basin should be within the oil window.

Recruitment patterns of postlarvae immigrating into mangrove nursery areas of five major estuaries around the south-eastern Gulf of Carpentaria, as well as juveniles emigrating offshore into coastal waters, were compared for the banana prawn, Penaeus merguiensis, from September 1978 to March 1979.. Although considerable variability was observed among rivers, some basic recruitment patterns were discernible. Recruitment of postlarvae tended to follow a 28-day cycle with increased immigration on alternate spring tides. Variability between rivers in the number of resident juvenile prawns at any one sampling time resulted mainly from differences in the relative magnitude of postlarval settlement from these monthly cohorts. After the first heavy rainfall of the monsoon season, the lower reaches of rivers with !xger catchment areas a!! ran fresh, setthg up a physica! barrier to further past larval immigration. In contrast, post larval immigration continued throughout the study period in the river with the smallest catchment. There was a trend for more successful immigration earlier in the more northern rivers. Offshore emigration was influenced by rainfall, tide height and number of resident juvenile prawns at the time of emigration. The relative importance of these three factors differed among rivers, depending on the timing of rainfall in relation to the timing of juvenile population changes and the degree of flooding. These local differences in the timing of emigration of juveniles could be detected in the abundance and size of adolescent prawns in the offshore coastal area of the south-eastern Gulf which in turn influenced the size composition of prawns available to the commercial fishery.

L. calcarifer undergoes a prolonged breeding season that starts just before the summer monsoon. In Van Diemen Gulf the fish spawned in 1978-1979 from September to February, and in the Gulf of Carpentaria in 1979-1980 from November to March. Spawning was localized, occurring near river mouths on coastal mudflats or upstream in rivers where salinities were relatively high (30). There was no evidence for coastal migrations to specific spawning grounds. Juveniles from 9 mm total length entered, on the high spring tides, supralittoral habitats such as tidal pools, gutters, swamps and low- lying inundated coastal floodplain for use as nursery areas.

The pattern of movements of the tiger prawns P. esculentus and P. semisulcatus in the western Gulf of Carpentaria, is described on the basis of three tagging experiments carried out in 1981. During these experiments, 13 180 individuals of P. esculentus and 6812 individuals of P. semisulcatus were tagged with Floy streamer tags and released in waters adjacent to Groote Eylandt. The first of the experiments was conducted in inshore waters north of Groote Eylandt in February. The second and third experiments were conducted in waters farther offshore north and south of Groote Eylandt in May-June. Logbook fishing-effort data were used in the interpretation of movement patterns from the geographic distribution of recaptures. From the February releases north of Groote Eylandt, both species showed movement offshore although, as a result of different directions of movement, the two species exhibited some degree of separation in the offshore fishery. The distribution of P. esculentus recaptures was generally east of the release area whereas that of P. semisulcatus was north-east. The distribution of recaptures from the June releases north of Groote Eylandt showed much less movement and separation of species than for those released inshore in February. Similarly, the recaptures from the May-June releases south of Groote Eylandt showed relatively little movement. With a few notable exceptions;these prawns exhibited only a gradual dispersal from the area of release.

A new species of skink in the sand-swimming genus Lerista is described from two islands on the western side of the Gulf of Carpentaria, Australia. In lacking forelimbs and in having only a monodactyl or stylar hindlimb the new species is one of the six most reduced-limbed mem- bers of the genus. It appears to be most closely related to L. praepedita from far western Australia.

Recent observations of satellite-tracked drogued buoys in the Gulf of Carpentaria indicate a slow, clockwise mean circulation, which appears to be a permanent feature in the Gulf. Residual currents, derived from 3-5 months of recent current-meter observations, and from a numerical tidal model of the Gulf, are compared with the motions deduced from the buoys. North-west monsoon winds and density-induced currents enhance the clockwise circulation. However, application of the south-east trade-wind stress at neap tides drives a counterclockwise circulation, and at spring tides, a weak clockwise circulation. The large annual variation of mean sea level in the Gulf is a maximum in the south-east corner (75 cm): 70% of this can be accounted for by the effects of winds, atmospheric pressure, and steric variations, all of which exhibit marked seasonality, are approximately in phase, and are thus additive.

Crab I. in north Queensland supports a large all-year nesting population of Chelonia depressa and occasional nesting by Eretmochelys imbricata and Lepidochelys olivacea. Measurements of nesting turtles, eggs, nests, hatchlings are summarized. A preliminary model of C. depressa egg and hatchling mortality gave values for the proportion of eggs producing hatchlings into the sea ranging from 48% to 89% in different months. The nesting distribution of C. depressa is reviewed and Crab I. identified as its most significant breeding beach.

Sea snakes from the eastern Gulf of Carpentaria were regularly sampled from April to December 1976. Thirteen species of snakes were represented among the 341 individuals collected. Over 61 % of these were of one species, Lapemis hardwickii, and none of the other 12 species represented more than 10% of the catch. There were several major differences between our sample and earlier studies in the Gulf and none of these differences could be accounted for. We speculate that earlier studies may have been due to biased samples or to real changes in sea snake distribution and abundance patterns caused by fishing.

The seagrass communities of the Gulf of Carpentaria were mapped and classified in terms of their species composition, their areas of shoot surface and their above-ground biomass. A total of 906.4 km2 of seagrass habitat, fringing 671.1 km of coastline was identified, mapped and sampled. Eleven seagrass species (approx. 20% of all known species) were recorded. In all, 74% of the seagrass communities occur along open coastlines and are characterized by depth-zoned species distributions (intertidally and subtidally), variable shoot-surface areas and variable above-ground biomass values. Each zone was dominated bfone or two of the following species: Syringodium isoetifolium, Cymodocea serrulata, Halodule uninervis, Halophila ovalis and Halophila spinulosa. Of the remaining seagrass, 10% occurred on reef flats in mixed-species communities, !3% in a regionally restricted monospecific community of Halodule uninervis, and 4% in communities dominated by Enhalus acoroides.

E. ingens was found on about 1% of banana prawns, P. merguiensis, from the south-eastem corner of the Gulf of Carpentaria, but was absent from prawns taken north of the Staaten River. Prawns evidently became infected when they were juveniles in the estuaries, as there was a strong correlation between parasite length and host length and no immature parasites were found on adult prawns. The bopyrid is apparently long-lived and thus makes a convenient biological marker for prawns coming from the south-eastem estuaries. Extensive migrations of adolescent banana prawns appeared unlikely as nursery rivers contributed prawns to adjacent offshore grounds.

IN May of last year (1908) I had the good fortune to meet the Bishop of North Queensland (Dr. Frodsham) at Liverpool, and he gave me in conversation some valuable information as to the native Australian beliefs and customs based on his personal knowledge of the aborigines. He told me that he had travelled among the Arunta as well as among various North Queensland tribes, and he asked me whether I was aware that the Australian aborigines do not believe children to be the fruit of the intercourse of the sexes. His Lordship informed me that this incredulity is not limited to the Arunta, but is shared by all the North Queensland tribes with which he is acquainted, and he added that it forms a fact which has to be reckoned with in the introduction of a higher standard of sexual morality among the aborigines, for they do not naturally accept the true explanation of conception and childbirth even after their admission into mission stations. The Bishop also referred to a form of communal or group marriage which he believes to be practised among aboriginal tribes he has visited on the western side of the Gulf of Carpentaria, but, unfortunately, I had not time to obtain particulars from him on this subject.

Charts of the distribution of salinity, temperature, inorganic phosphate, nitrate nitrogen, oxygen, and particulate organic phosphorus, for the eastern Arafura Sea and Gulf of Carpentaria in August 1964 are presented. Interrelationships of these properties show that at least three water masses were identifiable in this month. Two were very low in nutrients (phosphate less than 0.20, nitrate less than 1.0 �g-atom/l) but differed in salinity (less than 33.00‰ and greater than 35.50‰). The third was high in nutrients (phosphate greater than 1.40, nitrate greater than 17 �g-atom/l) and had salinities between 33.80 and 34.70‰. The high nutrient water mass was derived from Banda Sea slope water at around 100-150 m, wlth its nutrients increased subsequently by biological action. The other two water masses were formed in the coastal region of West Irian and the Coral Sea. High surface oxygen saturation (139%) and accumulation of organic phosphorus in near-bottom waters of the eastern Arafura Sea were the result of an uplift of Banda slope waters, much earlier in the year than August. In the Gulf of Carpentaria, the August salinity temperature characteristics were formed by the southward drift along the eastern margin of Coral Sea waters, which increased in salinity and decreased in temperature by evaporation. Low salinity water of the previous summer occurred in August, only in the north-west of the gulf.

The hydrological properties of five major surface zones in the south-east Indian Ocean are discussed. These are the Subantarctic, the Subtropical, the Tropical, and two transition zones separating them. Two regions of upwelled water are described. The Subantarctic Zone lies south of the 35.00‰ isohaline and has an inorganic phosphate P content hetween 0.30 and 0.60 μg-atom/l. The Subtropical Zone lies between the 35.80‰ isohaline in the south and the 35.50‰ isohaline in the north and has an inorganic phosphate P content between 0.10 and 0.15 μg-atom/l. For this zone waters with salinity greater than 36.00‰ in the south-east Indian Ocean and greater than 36.30‰ in the Great Australian Bight constitute the major source regions of high salinity water. Both of these, but particularly the latter, are very low in phosphate. The Tropical Zone lies between the south equatorial salinity minimum (34.26‰) and the Monsoon Current (salinity 34.10‰), at about 2°N. Four hydrological regions have been distinguished in this zone. From north to south these are (1) the Monsoon Current with low salinity (34.10‰), high temperature (28.9°C), and low phosphate P (0.12 μg-atom/l.); (2) the Equatorial Counter-Current with moderately high salinity (34.50‰), high temperature (28.7°C), and high phosphate (0.21 μg-atom/l.); (3) the South Equatorial Divergence with moderately high salinity (34.48-34.55‰), moderate temperature (26-27.5°C), and high phosphate (0.20-0.30 °g-atom/l.); (4) the South Equatorial Current with waters at its core of salinity 34.26-34.31‰, temperature 26-27°C, and phosphate 0.20-0.25 μg-atom/l. In the transitional zone between the Subantarctic and the Subtropical Zones salinity changes rapidly with latitude. The centre of the zone is marked by the position of the 35.40‰ isohaline. The centre of the transitional zone between the Subtropical and the Tropical Zones is marked by the position of the 34.90‰ isohaline. Two regions of upwelled water were found: (1) along the shelf and elope regions of north-west Australia south of Timor, distinguished by high phosphate P (0.20-0.30 μg-atom/l.) and a maximum upwelling occurring before September; (2) along the shelf and slope region of the eastern Arafura Sea, from the Aroe Islands to the Gulf of Carpentaria,, where the highest surface phosphate values in oceanic waters of Australia were found (0.66 μg-atom/l.), and a maximum upwelling probably occurring before August.

In this memoir Mr. Jukes gave a short abstract of all the information collected by various travellers regarding the Australian continent, including his own observations. The eastern coast is occupied by a great range of high land, appearing like a continuous chain of mountains when seen from the sea, and rising in several places to 5000 feet or more above the sea-level. This chain has an axis of granite, with occasional large masses of greenstone, basalt and other igneous rocks. It is flanked on both sides by thick beds of palaeozoic formations, chiefly sandstone, but also containing limestone and coal. In the northern portion of the chain Dr. Leichardt found similar formations, and especially trap and granite near the Burdekin river. In the Port Philip district there are similar igneous rocks, and on the coast tertiary formations, which Mr. Jukes found resting on the edges of upturned palaeozoic beds. In West Australia the Darling range consists of granite below, covered by metamorphic rocks; and between it and the sea is a plain composed of tertiary beds. In the colony of North Australia there is a great sandstone plateau, rising about 1800 feet above the sea, and probably of palaeozoic age; whilst on the immediate shore and round the Gulf of Carpentaria are beds supposed to belong to the tertiary period. Similar formations constitute the substratum of the central desert, in which Capt. Sturt was compelled to turn, when half-way to the Gulf of Carpentaria, from the southern coast.

The Leichhardt-Gilbert area of northwest Queensland comprises four physiographic divisions, the Isa Highlands in the west, the Carpentaria and Inland plains in the center, and the Einasleigh Uplands in the east. A plain of erosion of late Tertiary-Quaternary age occupies large areas of the central plains, and remnants of erosional plains of early to middle Tertiary and pre-middle Mesozoic ages are prominent elements of the present land surface in the two upland areas. The three plains are considered to be subaerial and polycyclic in origin; a geologically recent emergence has caused renewed downcutting near the coast. Except where the lithology is conducive in certain localities on the youngest plain of erosion no evidence suggests that pediplanation is or has been the dominant process molding the landscape. However, the writer agrees with the theses of L. C. King that in the Leichhardt-Gilbert area there are three major surfaces of erosion and that laterization occurred earlier than previously postulated.