Mangareva, along with several other islets from the ‘Gambier almost-atoll’, are residual parts of a past volcano pertaining to the Gambier Archipelago (South Pacific). The islets are surrounded by a vast lagoon delineated by a partly subsided coral barrier reef. Benthic diatom assemblages from these marine environments included, among others, Achnanthales and species of the elusive genus Olifantiella, with Cocconeis, Amphicocconeis and Planothidium as the most diverse genera. Hao, a large Tuamotu atoll 830 km northwest of Gambier, was sampled at about the same time. A Non-Metric Multidimensional Scaling applied to data from Mangareva and Hao highlighted groups of locations and biotopes. A PERMANOVA analysis was applied to test for the effect of both biotope and location on the differences in community composition. The geologic history of the reef structure (from volcanic and silicic environments surrounded by a reef in Mangareva, to typically coral and calcic environments in Hao, with a much older geologic history) may have played a role in the degree of colonization by benthic diatoms. Some taxa and morphs were only observed in Mangareva, whereas some others, present in north Tuamotu, were apparently absent in Gambier, permitting some remarks on tropical diatom biogeography and possible endemism, taking into account that the Gambier sector is located to the south in the South Equatorial Current, and thus largely removed from the influence of other Polynesian islands.

Since the 1970s only one Ruellia species (Ruellia blechum L.) has been reported from Guam. A few recently introduced Ruellia species have become established over the island but have not been identified or documented. In this paper, we add four Ruellia species (Ruellia prostrata Poir., Ruellia simplex C.Wright, Ruellia tuberosa L., and Ruellia ciliatiflora Hook.) to Guam’s weed list, most of which appear highly aggressive in our urban environment. These Guam Ruellia can be differentiated using the key and photos we provide. We also briefly discuss mechanisms that could contribute to the “success” of these plants.

Information on the cetacean fauna found in the nearshore waters of Pagan Island, Commonwealth of the Northern Mariana Islands (CNMI), are reported for the first time. In August 2013 visual and acoustic surveys in the nearshore (< 5.6 kilometers; km [3 nautical miles; nmi]) waters surrounding Pagan Island collected data on cetacean occurrence using line transect (visual and towed acoustic), photo-identification, and moored sonobuoy (passive acoustic monitoring) methods. Three species of cetaceans were visually confirmed at Pagan Island: spinner dolphin (Stenella longirostris), common bottlenose dolphin (Tursiops truncatus), and Cuvier’s beaked whale (Ziphius cavirostris). Acoustic encounters included spinner, common bottlenose dolphin, Cuvier’s beaked whale, Blainville’s beaked whale (Mesoplodon densirostris) and sperm whale (Physeter macrocephalus). The moderately high detection rates of beaked whale species at Pagan can likely be explained by the steep bathymetry and deep-water habitat close to shore. Sperm whales were detected on the acoustic recordings from nighttime moored sonobuoys, indicating their presence within 37 km (20 nmi) of Pagan. On 10 of the 11 research days at Pagan, dolphins were visually or acoustically encountered in the study area, during day and nighttime sampling periods. Photo-identification yielded re-sights of individual spinner dolphins at Pagan Island on successive days. A potential breeding population of bottlenose dolphins was documented at Pagan based on the presence of three calves, including two neonates. The visual and acoustic detection rates for dolphins were relatively low. Caution should be used in interpreting any of the detection rate results due to the small sample sizes. The density or abundance of cetaceans in the study area could not be estimated because of small sample sizes. If future surveys are conducted using similar methods the line transect survey data could be pooled to estimate density and abundance.

The Mariana skink (Emoia slevini) is a federally endangered lizard whose ecology is poorly known. It was considered extirpated on Islan Dåno’, Guam, but was rediscovered there in 2012. Here, we delineate its distribution as of 2017-2018, describe its habitat use, and evaluate density of the more common Pacific blue-tailed skink (Emoia caeruleocauda) as an indicator of habitat suitable for E. slevini. We conducted visual encounter surveys across Dåno’ followed by intense surveys for E. slevini throughout all areas of the forest type where it was first detected. Geospatial analysis showed a limited distribution of individuals in area constituting 6.2–12.8% of the island’s total area. Overlay of E. slevini locations (n = 199) on a digital vegetation/land cover layer revealed that, under conditions of complete spatial randomness, it was over-represented in Coconut Forest, followed by Mixed Introduced Forest, and underrepresented in three other vegetation classifications. This analysis, when conducted for E. caeruleocauda (n = 296), revealed a similar pattern. An investigation into the association of E. slevini locations with individual coconut palms digitized from aerial imagery revealed a positive clustering of individual E. slevini at the 99% confidence level nested within the 90-99% level for palms, indicating the importance of this tree species on a landscape scale. Analysis of perch types showed downed palm debris (dead fronds and coconuts) were the most frequently used perch types, demonstrating the importance of this tree at the microhabitat scale. Although widespread on Dåno’ in the 1950’s, Coconut Forest is presently the most limited forest type constituting only 1.2% of the island’s total area. A comparison of hot spot clustering for the two lizard species revealed that clustering for E. slevini was nested within the 90-99% level for E. caeruleocauda, further supporting the feasibility of using density of E. caeruleocauda as an indicator of habitat favorable for E. slevini.