Coral blocks were used as settlement surfaces to study the epilithc algal community (EAC) in the presence and short-term absence of large, herb~vorous grazers on the central Great Bamer Reef (GBR). Grazers were excluded by cages at study sites on 2 lnshore, 2 midshelf and 2 outershelf reef. Of 82 algal taxa recorded, 81 were small filamentous or fleshy 'algal turfs' C o m p o s~t~o n and abundance of algae differed greatly between caged and uncaged substrata, w~t h differences more pronounced on mid-and outershelf reefs than on lnshore reefs. Caged surfaces on lnshore reefs were dominated by Sphacelaria novae-hollandiae (Phaecophyta), an assemblage of Oscillatoriaceae (Cyanophyta), Acetabularia calyculus (Chlorophyta), Cerarniurn flaccidurn, Taeniorna nanum and Polysiphonia infestans (all Rhodophyta) and Cladophora fasiculans (Chlorophyta) Caged surfaces on m~d -and outershelf reefs were dominated by Enterornorpha clathrata (Chlorophyta), P. ~nfestans, PolySiphonla sphaerocarpa, C. flaccidurn, Herposlphonla secunda f tenella and Lobophora variegata (all Rhodophyta), S. novaehollandae, assemblages of Osc~llatonaceae and Nostocaceae (Cyanophyta), Glfforda mitchellae and l Pseudopringsheimla (both Chlorophyta) These species are likely to account for a large proportion of algal product~on and food of large, herb~vorous grazers on these reefs. Uncaged surfaces on inshore reefs were dominated by S. novae-hollandiae, an assemblage of Oscillatonaceae, G. rnitchellae and P. sphaerocarpa; those on m d -and outershelf reefs by the crustose coralhne alga Porolithon onkodes, assemblages of Nostocaceae and Oscillatoriaceae, ? Pseudopringsheirnia, Polysiphonia scopulorurn and Ceramluni punctiforrne. Locat~on on the continental shelf accounted for the major proportion of variability In the data due malnly to the dstribution of P onkodes. When thls species was om~tted from the analysis, the effect of excludng grazers In combination ulth differences between lnshore and mid-/outershelf reefs accounted for the major proportion of vanability. In the latter analysis, locatlon was a far less important component of var~ability In EAC omposit~on than was presence or absence of grazers, because of large between-reef (I.e. within-locahon) variance in the abundance of many species. Grazers have a greater lmpact on compos~tion and abundance of the EAC on mid-and outershelf reefs than on inshore reefs, in agreement with previous findings that herbivorous, grazing fishes are significantly less abundant on ~n s h o r e than on mid-and outershelf reefs.

The principal genera of plants consumed by green turtles in Torres Strait were Hypnea, Laurencia, Vidalia, Sargassum and Thalassia. Although some selectivity for soft red algae was detected, no evidence was found to support the theory that turtles feed on algae only when seagrasses are unavailable, but nor do algae appear to confer a nutritional advantage on those animals that eat them. Geographical variation in green turtle diets appears to be determined by the relative availability of seagrass and algae and the structure of the local herbivore community.

No abstract

documented clinical visit in the year prior to the pregnancy. Of 177 visits, primary care was the most common setting (46.5%). A total of 100.0% of gynecologists, 97.1% of primary care providers and 40.0% of emergency department providers documented reproductive topics. Women with social risk factors (e.g., history of abuse, homelessness, or exchange of sex for drugs or money) were less likely than other women to have such documentation (70.0% vs. 96.3%, p=.02), and women who had psychiatric comorbidities were less likely than women who did not to receive this counseling (47.4% vs. 77.4%, p=.03). Conclusions: We identified an HIV-positive population of women who experienced an unintended pregnancy. Emergency department visits may represent an opportunity to offer more reproductive counseling, perhaps by integrating an EMR tool or improving access to social workers. Providers were less likely to document reproductive topics for patients with social risk factors or psychiatric comorbidities, suggesting that these factors may contribute to reproductive counseling disparities.